Today is International Labour Day.

A day that started as a tribute to workers who fought for basic dignity, fair hours, and safe conditions.

A day that in 2026 carries a particular weight, because a lot of people are heading into the long weekend quietly worried about whether there will still be a job waiting for them when the world catches its next breath.

That worry deserves to be taken seriously, not dismissed with optimistic hand-waving about new opportunities and inevitable progress.

So let’s be honest about what is actually happening, because the full picture is more complicated, and ultimately more hopeful, than most of the current discourse suggests.

## **The Collective Action Problem Nobody Wants to Name**

El Salvador’s President [Nayib Bukele](https://x.com/nayibbukele/status/2043185432972722445) posted something on X last month that cut through a lot of noise.

![Source: X](/api/media/elephantstripes-prod/cms/originals/blog-inline/ab2d15836626eb840fd7e00ec900425f97115ccc4838ea582c3ed4569dca9dd1.avif "Source: X")

That is an unusually clear-eyed observation from a sitting head of government, and it names the structural risk that most corporate communication carefully avoids. Individual companies are making rational decisions. The aggregate result of those decisions is potentially irrational at the system level.

![Challenger Report Annual Job Cut Announcements 2015–2025 (Source: Challenger, Gray & Christmas, 2026).](https://miro.medium.com/v2/resize:fit:1400/0*iI-q68S-uEVSEEgC.png "Challenger Report Annual Job Cut Announcements 2015–2025 (Source: Challenger, Gray & Christmas, 2026).")

[Challenger, Gray & Christmas](https://www.challengergray.com/blog/2025-year-end-challenger-report-highest-q4-layoffs-since-2008-lowest-ytd-hiring-since-2010/) reported that AI was blamed for 50,000 layoffs in 2025. The tech sector is the most visible example, but the pattern is spreading into marketing, legal support, customer operations, and financial services. Real people are losing real jobs, and the downstream effects on spending, confidence, and community tax bases are real too.

Sam Altman has said publicly that some companies are using AI as a convenient scapegoat for workforce decisions that had more to do with pandemic over-hiring and poor financial management than genuine automation.

There is some truth to that. Corrections have been underway since 2022, and most have not been associated with AI adoption, which really began to affect coding in 2023 and 2024. In other words, the layoffs began for financial reasons.

But that framing is also a convenient deflection. **The jobs are still gone**. The scapegoat versus genuine cause distinction matters less to the person updating their resume at midnight than it does in a policy debate.

## **The “Tech Hiring Is Dead” Narrative**

@[youtube](https://www.youtube.com/watch?v=0j1HqEEDThc)

Dario Amodei at Anthropic has made statements about AI wiping out white-collar jobs and coding roles within the next few years. He said it plainly, and the comments rippled through the tech community in ways that reinforced a growing narrative: that a career in software development is now a bad bet.

I do not think that holds up against the actual data.

The [Indeed AI Tracker](https://www.hiringlab.org/2026/01/22/january-labor-market-update-jobs-mentioning-ai-are-growing-amid-broader-hiring-weakness/) reached a high of 4.2% at the end of 2025. Job postings mentioning AI have grown consistently since mid-2023. Employer job postings related to AI jumped 117% between 2024 and 2025.

LinkedIn data tells the same story. AI has already added more than [1.3 million new roles](https://www.weforum.org/stories/2026/01/ai-has-already-added-1-3-million-new-jobs-according-to-linkedin-data/?gad_source=1&gad_campaignid=22228224717&gbraid=0AAAAAoVy5F4JlvyK4rUqmaS-YvSZPOeID&gclid=CjwKCAjw-8vPBhBbEiwAoA39Woe6mQZX1_buFGUtI9YbtRr40-az4skZCiUwUA81om3E-BfNLAXS9RoCNTEQAvD_BwE), including AI Engineers, Forward-Deployed Engineers, and Data Annotators. More than 600,000 new AI-enabled data centre jobs have been created.

[The US Bureau of Labour Statistics](https://www.bls.gov/ooh/computer-and-information-technology/) projects computer and information technology occupations to grow at nearly three times the average rate for all jobs through 2034. Professionals with AI expertise earn 56% more on average than peers without it.

This is not a hiring market where tech is dead. It is a hiring market where the skills being rewarded have shifted dramatically, and where people who have not made that shift yet are experiencing real pain.

The distinction matters. “Tech hiring is dead” is a counsel of despair that leads people to exit a field where opportunity is actually growing. “Tech hiring has shifted and requires different skills” is uncomfortable but actionable.

For workers, developing and highlighting relevant AI skills may be the key to landing a job in 2026, particularly in occupations with otherwise muted hiring activity. That is not cheerful messaging. But it is accurate, and accurate is more useful than comfortable.

## **What the Broader Market Actually Looks Like**

![Photo by Srinivasan Venkataraman on Unsplash](https://miro.medium.com/v2/resize:fit:1400/0*Dm6u-bhVbOOpEsiB "Photo by Srinivasan Venkataraman on Unsplash")

The overall picture is one of genuine contradiction, which is why the conversation is so difficult.

Companies are laying off staff, insisting AI will “do more with less,” yet they haven’t found ways to deploy AI at scale. Recruiters say entry-level pathways are narrowing, but critical roles remain hard to fill.

Just over half of people say they’re job hunting in 2026, while nearly 80% feel unprepared to find a new job. That gap between the number of people searching and the number who feel equipped is where a lot of the anxiety is living.

Recent college graduates are bearing a disproportionate share of the disruption. Those aged 22 to 27 saw an unemployment rate of 5.6% at the end of last year, compared to the overall rate of 4.2% (The Federal Reserve Bank of New York, [2026](https://www.newyorkfed.org/research/college-labor-market#--:explore:unemployment)). Many of these graduates were headed into entry-level roles at law firms, research organisations, tech companies, and public sector agencies, precisely the environments now adopting AI to handle the tasks that basic training prepared those graduates to perform.

[The World Economic Forum](https://www.weforum.org/press/2025/01/future-of-jobs-report-2025-78-million-new-job-opportunities-by-2030-but-urgent-upskilling-needed-to-prepare-workforces/) estimates that AI could eliminate 92 million jobs by 2030 while creating 170 million new ones, a net gain of 78 million. That number is meaningless to someone who lost their job in accounting in 2025 and cannot afford to retrain for an AI engineering role by 2030. The aggregate optimism and the individual pain are both real at the same time.

## **Why This Prompted Us to Build Something**

At [Elephant Stripes](https://elephantstripes.ai/), my team and I have been watching this closely. Not from the outside, but as people who care about the practical reality of what workers are navigating right now.

The job seeker’s experience in 2026 is genuinely difficult in a specific way. The advice is abundant. The tools are everywhere. The templates are free and ubiquitous. And yet most people still have resumes that do not actually represent what they are capable of.

The standard AI resume approach makes this worse, not better. You paste your existing resume into a language model, it polishes the language and adds keywords, and you end up with a shinier version of the same document. Generic. Template-heavy. Optimised for an ATS screening tool that has seen ten thousand copies of the same format.

![](https://miro.medium.com/v2/resize:fit:1400/1*yqQR-REINmH1IeX8fFTUlQ.png)

That is why we built [**JobsLobster**](https://jobslobster.com/) differently.

JobsLobster does not just rearrange what you already have. It identifies what is missing. It asks you targeted questions about your actual skills, your specific contributions, and the outcomes you produced. And it builds something stronger from your answers, rather than from the limited version of yourself that ends up in a resume written in haste.

The differentiator is in the input. Most resume tools start with your resume. JobsLobster starts with you.

## **A Note on What Comes Next**

The transition happening right now is real and it is not finished. The disruption to entry-level work is real. The anxiety is legitimate. The companies using AI as convenient cover for decisions they made for other reasons are being less than honest about it.

And the future is not one in which technology simply eliminates work. Deloitte found that while 30% of organisations are exploring agentic AI options and 38% are piloting solutions, only 14% have solutions ready to deploy, and a mere 11% are actively using these systems in production. The wave is coming, but it is not a wall that hit overnight.

The workers who will navigate this best are the ones who are building AI fluency now, not as a panic response to headlines but as a steady, deliberate investment in the skills that the next decade of work will reward.

That is not naive optimism. It is what the data actually says.

Happy Labour Day. The work is changing. So are the people doing it.

\---

_Are you interested in getting early access to JobsLobster? Drop us a message_ [_here_](https://elephantstripes.ai/contact) _to stay updated!_

Labour Day in the Age of AI Layoffs - Tat Banerjeee

Labour Day in the Age of AI Layoffs

A Year 9 student from Korea, studying at an Australian school, wanted to celebrate her friend’s birthday. The request seemed straightforward enough: it was an overnight camping trip in a nearby national park.

For any teenager, this is the stuff of core memories. Campfires, stars, the particular magic of being outdoors with friends.

For her school, it was the beginning of a complex risk assessment that would test the limits of traditional communication.

How do you explain duty of care obligations across a language barrier? How do you ensure parents truly understand the risks their child might face? And how do you protect your school if something goes wrong? These questions loomed large, creating a sense of unease.

This is where most schools would have said no. Because it’s too complicated. Too risky. Too hard to ensure everyone truly understood what they were agreeing to.

But this school found a different way.

_This is a real story from one of our clients in the education sector._

## **When Permission Slips are Not Enough**

![Photo by Anna Amelina on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/5e9a1f6d0982e433c1e0cf4f5e95cb27898a9f577a3531117b095a439bf5e6cc.webp)

The student was a minor, and her parents lived in Korea. Under Australian duty of care obligations, the school bore responsibility for her welfare. They could not simply send home a permission form and hope for the best.

The campsite presented genuine risks. Emergency vehicle access was limited, with boat transport the only option between 9 PM and 6 AM. The bushland environment meant potential encounters with snakes and other wildlife. Water safety was a consideration. Adult supervision ratios needed to be carefully planned.

None of these risks were insurmountable. But the school's leadership knew that the parents needed to understand them fully before giving consent.

Not a cursory acknowledgment buried in fine print, but genuine comprehension of what their daughter might experience.

The problem was obvious: _how do you have that conversation across 8,000 kilometres and a significant language barrier?_

## **Finding A Way Through**

Traditional options felt inadequate.

Written translations could convey facts but not nuance. Phone calls with interpreters were difficult to schedule across time zones and left no reliable record. The school needed something that combined the immediacy of face-to-face conversation with the accessibility of real-time translation.

They found their solution in a [**VideoTranslatorAI’s**](https://agent.videotranslator.ai/) **video call tool** that provided live translation and on-screen transcription. The assistant principal could speak in English while the parents followed along with Korean text appearing in real time. Questions could be asked and answered immediately, with both parties confident they understood each other.

![English-Korean Multilingual Video Call](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/159d52d83898ad9a1180953037d26d556be006b23c774393d4dfd342da2a511a.webp)

_(Multilingual video call featuring AI-generated faces. Real clients' faces are not displayed to protect their privacy.)_

The conversation covered everything: supervision arrangements, emergency protocols, environmental hazards, and the precautions the school had put in place.

The parents asked thoughtful questions. They appreciated the detail. They gave their consent with full knowledge of what their daughter's adventure would involve.

And the school had something equally valuable: a documented record of the entire discussion, evidence that informed consent had been properly obtained.

The student went on her camping trip. She experienced Australian bushland, celebrated with friends, and returned safely with memories that will last a lifetime.

That’s a win for everyone!

## **The Bigger Picture**

![Photo by Mitchell Luo on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/bbf0db53ad57549c98065fd49cf6e98c8e6804570a5ff4231dc4fa7c4c022299.webp)

This wasn't just about approving a camping trip. It was about equity of opportunity for international students.

When language barriers make complex conversations too difficult, schools default to "no."

Not because the activity is genuinely unsafe, but because communicating risk and obtaining informed consent feels impossible across language differences.

The result? International students miss out on excursions, sporting events, cultural experiences, and opportunities that their English-speaking peers access easily. Not because their parents don't care, but because the school can't effectively communicate with them.

Beyond this situation, language barriers in education impact numerous touchpoints throughout the school year:

*   **Parent-teacher conferences**: Without effective communication, parents struggle to understand their child's academic progress, behaviour concerns, or additional support needs.
    
*   **Special education services**: Obtaining an Individualised Education Programme or 504 plan is complex even for English-speaking parents. When language barriers prevent parents from understanding their child's rights and navigating these processes, problems compound, leading to increased spending on interventions down the road.
    
*   **Emergency communications**: When urgent situations arise, schools can't afford translation delays. Parents need immediate, clear information they can act on.
    
*   **Permission and consent**: Any situation involving risk, medical treatment, or student welfare requires documented informed consent. Without clear communication, schools expose themselves to legal liability.
    
*   **Cultural integration**: International students and their families need to understand school policies, expectations, and cultural norms. When communication breaks down, misunderstandings create unnecessary conflict.
    

## **The Documentation Advantage**

In addition to improved communication, there is a practical benefit that school administrators increasingly value: **evidence**.

In an environment where institutions must demonstrate compliance with duty of care obligations, having a clear record of parent communications is invaluable.

Transcripts of video calls provide documentation that emails and phone notes simply cannot match.

The camping trip approval was not just a successful conversation. It was a legally defensible record of informed consent, captured in real time, with clear evidence that the parents understood and accepted the risks involved.

For schools navigating complex regulatory environments, this kind of documentation is not a luxury. It is a necessity.

![What Happens When Consent Gets Lost in Translation](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/61d3f6dfbd8da5cb0c5b22d678c1eff79abec1aba2df4c1f21111a4c68a328d1.webp)

_(Instant meeting transcript and summary by VideoTranslatorAI)_

## **A New Standard for Inclusive Education**

![Photo by Sam Balye on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/30e1dc76137f5dba327d82f23e35f63738d5e89def2ed829423ce7ef2c57a0d0.webp)

International education has always required cultural sensitivity and adaptability. But for too long, language barriers have created invisible walls between schools and families.

[VideoTranslatorAI’s](https://agent.videotranslator.ai/) real-time translation technology does not just solve a logistical problem. It signals something important to families: that their understanding matters, that their involvement is valued, and that distance and language will not diminish their connection to their child's education.

The Korean student who went camping learned something valuable about Australian culture that week. But her parents learned something too: that their daughter's school saw them as genuine partners in her education, not obstacles to be managed.

That shift in relationship is worth more than any single permission slip.

When schools communicate clearly with every family, students get opportunities they might otherwise miss. Real-time interpreting tools are making that clarity possible, one conversation at a time.

**_Want to improve communication with your international families?_** [_VideoTranslatorAI_](https://agent.videotranslator.ai) _offers multilingual video calls that enable real-time, face-to-face communication between educators and parents, regardless of language differences._

If you're interested in learning more, [try our product now](https://agent.videotranslator.ai/chat).

## Read more

*   [Sales Didn’t Kill My Deals. Silence Did.](https://videotranslator.ai/blog/sales-didnt-kill-my-deals-silence-did/)
    
*   [This AI Should Never Have Been Built](https://videotranslator.ai/blog/this-ai-should-never-have-been-build/)
    
*   [I Built Software Wrong for 7 Years. You Shouldn’t Do The Same.](https://videotranslator.ai/blog/i-built-software-wrong-for-7-years-you-shouldnt-do-the-same/)

What Happens When Consent Gets Lost in Translation

A real case that exposed a hidden international school risk

You just finished a brilliant product demo. The prospect loved it. They asked smart questions, nodded at all the right moments, and promised to "circle back next week."

Then you spotted a critical bug in production.

Three weeks later, you remember the demo while brushing your teeth. You scramble to find the email thread. The prospect has gone cold. Another warm lead, lost to the chaos of founder life.

This kept happening to me. Not because I lacked ambition or sales skills, but because I kept choosing to ship features over sending follow-ups. 

And I wrote this article to tell you how I fixed this problem. 

## The Founder's Impossible Choice

![Photo by ThisisEngineering on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/144ef7f1933d752669ed763ee5b15e193754d6afd6a765ca43926b6ac2761ed4.webp)

As a small startup founder, I discovered that cold outreach and product demos were my most effective marketing channels. Nothing beats a live conversation where you can address objections in real time and show your product solving real problems.

But here's what nobody tells you about founder-led sales: the work never stops competing with itself.

Every hour I spent nurturing leads was an hour I wasn't fixing bugs. Every afternoon I dedicated to follow-up emails was an afternoon I wasn't shipping features. 

My brain constantly toggled between "build mode" and "sell mode," and **build mode almost always won.**

The result? Warm leads turned cold. Interested prospects forgot about me. Deals died in silence.

My goal is modest but meaningful: 100 loyal users by the end of 2026\. 

But I realised I would never reach it if I kept losing prospects to my own distraction.

## The Hidden Momentum Problem

When I realised I was losing leads, my first instinct was to look for a CRM solution. Surely there's a tool for this, right?

There are. Hundreds of them. However, they all miss the point.

Traditional CRMs are designed for sales teams with pipelines, stages, and conversion funnels. They're built for people whose full-time job is sales. They assume you have time to update dashboards, manage pipelines, and maintain detailed contact records.

As a founder, I don't have that time. I'm not a full-time salesperson. I'm a builder who occasionally needs to sell what I've built.

What I actually needed was simpler: 

**Tell me who I should follow up with today, and remind me what I already did.**

That's it. No complex pipelines. No elaborate stages. Just a daily list and a memory aid.

## A Nature Documentary that Changed My Thinking

Last Christmas, I found myself watching a nature documentary about honey badgers. 

These small African mammals have a reputation for fearless persistence. They chase prey twice their size. They raid beehives despite getting stung hundreds of times. They never give up.

Something clicked while watching. The best salespeople aren't aggressive or pushy.

They're persistent. They show up consistently, follow up reliably, and never let promising conversations fade away.

I decided to build a tool that would help founders adopt this mindset. I called it ***Honey Badger Sales.***

![Photo by Vincent van Zalinge on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/5847aadc6e943722ae84919c3a3a204b28a461a0520a1f777cf5d867f74342e3.webp)

## How The Tool Actually Works

I built Honey Badger Sales to solve one problem exceptionally well: **making sure warm leads never go cold.**

The workflow is simple. First, I create a new contact list or import from my existing CRM data. No complex migration required.

![Add Company propspect](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/8fc349f90b78220f5895fd5be97a899ebeeb3f9c4452e755e0078d507cc8d123.webp)
*(Adding a prospect contact on Honey Badger Sales)*

From there, I can draft and send outreach emails directly within the app. Everything stays connected to the contact record, so I always have context at my fingertips.

![Write or draft email](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/834129e0af07fb817238fd47029867f890acc0a6235dd6ed24db991a9a45e712.webp)
*(Send or draft an email outreach easily!)*

The homepage shows my daily focus list: contacts who need follow-up today. Each contact clearly displays whether they're due for a first follow-up, second follow-up, or fifth. I can also see the last thing I discussed with my prospect and what I planned to say next.

![Honey badger homepage](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/ebc0b374b736890f1ccbbf9426e9f971651758288866682bda7d71adf8c4ad55.webp)

That's it. No fancy pipelines. No overwhelming dashboards. Just a daily list and the memory you need to pick up where you left off.

## What I've Learned Building This

Creating Honey Badger Sales taught me something important about founder-led sales: consistency beats intensity every time.

I used to batch my sales work into marathon sessions. I would spend an entire day doing outreach, then disappear into code for two weeks. This created a boom-and-bust pattern that confused prospects and killed deals.

Now I spend 30 minutes each morning working through my follow-up list. It's not glamorous. But those 30 minutes, repeated daily, generate more results than sporadic bursts of intense effort.

The honey badger doesn't make one dramatic attack and hope for the best. It keeps showing up, day after day, until it gets what it came for.

## Your Leads are Waiting

![Photo by Lukas Blazek on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/ef41613426c323b951ca997939557c013409f198769519be04d867f144d60b42.webp)

If you're a founder struggling to balance building and selling, you're not alone. The pull toward product work is real and powerful. Shipping features feels productive in a way that sending follow-up emails simply doesn't.

But those warm leads represent real people who showed genuine interest in what you're building. They deserve a response. Your startup deserves the revenue.

Sometimes the best tool isn't the most sophisticated one. Sometimes it's the one that simply reminds you to show up.

Be the honey badger. Never let a warm lead go cold\!

**For now, Honey Badger Sales is something I built for myself. I'm still tweaking it, learning what works, and using it daily to manage my own outreach.*

*But if the problem I've described sounds familiar, I'd genuinely like to connect. Leave a comment or reach out to us by clicking here.*

## Read more
- [This AI Should Never Have Been Built](https://videotranslator.ai/blog/this-ai-should-never-have-been-build/)
-  [I Built Software Wrong for 7 Years. You Shouldn’t Do The Same.](https://videotranslator.ai/blog/i-built-software-wrong-for-7-years-you-shouldnt-do-the-same/)
- [You Sound Like ChatGPT Now](https://videotranslator.ai/blog/you-sound-like-chatgpt-now/)



Read Sales Didn’t Kill My Deals. Silence Did. and explore practical insights from VideoTranslatorAI on multilingual communication, interpretation, and AI-supported workflows.

Sales Didn’t Kill My Deals. Silence Did.

Open X (Twitter) right now and watch people become other people.

Not through makeup or costume. But through AI that maps their movements onto someone else's face with unsettling precision. 

The technology tracks every micro-expression, every head tilt, every blink. The result moves like a human because it is a human, just wearing digital skin that doesn't belong to them.

![Screenshot 2026-01-21 at 00.14.23.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/5f7bf76cfd721b22c1061dfa48609c35d23a2e9a2a940a18b18916076af87192.webp)
_(Posted by [@venturetwins](https://x.com/venturetwins/status/2011285029541077033) on X)_

Leonardo DiCaprio, Sydney Sweeney, your favourite actor, your least favourite politician, your ex. Anyone whose face exists in a training dataset.

The demos are technically impressive. The implications are horrifying. 

And I say this as someone who builds AI for a living.

## The Question We Stopped Asking
![Photo by Simone Secci on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/7f4dd031df294e1c596ffca3c2f2ef48789ba538ba502d244b5e88b565002eab.webp)

I run a company that develops AI-powered meeting tools. Real-time transcription. Live translation across languages. Intelligent summarisation. Every feature we ship starts with the same question: **what problem does this solve?**

It's not a complicated question. But somewhere in the race to build the most impressive thing, a lot of AI developers stopped asking it.

Now, let me ask you a question. 

**What problem does face-mimicking AI solve?** I dare you to name one legitimate application.

**Entertainment?** Perhaps. But we already have professional CGI, deepfakes with consent, and actors who can, you know, act.

**Marketing?** Sure, until Leonardo DiCaprio's lawyers come knocking because you used his face to sell protein powder without permission.

**Education?** I'm struggling to imagine the scenario where transforming yourself into Scarlett Johansson helps students learn algebra.

**Creative expression?** There's an argument here, but it falls apart the moment you realise most of these tools are being used to impersonate real, living people who never consented.

I've spent fifteen minutes on this exercise. Then an hour. Then I asked my team. We came up empty.

The charitable interpretation is that this technology is a solution searching for a problem. 

The uncharitable, and I suspect more accurate, interpretation is that the creators know exactly what it's for. They're just not saying it out loud

## What It's Actually Being Used For

Let me tell you what I'm seeing on my timeline.

People turning themselves into celebrities for clout. 
![Screenshot 2026-01-21 at 00.13.29.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/c263f1d5a4435955192274ff95a5b93172c3052e0c21c88df9fe6b446705b62e.webp)
_(Posted by [@EHuanglu](https://x.com/EHuanglu/status/2011812530637959629) on X)_

People creating "erotic" content using the faces and bodies of figures who never agreed to this. 
![Screenshot 2026-01-21 at 00.12.18.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/d4c5919e8a570395d59483af8f111c4772978696a37e6764a177ba80fbe8205d.webp)
_(Posted by [@mhdfaran](https://x.com/mhdfaran/status/2013257382345015320) on X)_

People testing just how convincing they can make a fake video of a real person.

This isn't hypothetical harm. It's happening right now, in public, with thousands of likes and shares.

The erotic applications alone should give us pause. We've spent years grappling with non-consensual intimate imagery. 

We've passed laws. We've built detection tools. We've had painful public conversations about consent and dignity.

And now we're handing people a tool that makes all of that exponentially easier to produce and harder to detect.

If you're building technology and you can't articulate a beneficial use case, but you can immediately see the harmful ones, that's not a feature gap. That's a warning sign.

## My Three Core Concerns

With this face-swapping technology, I see three glaring issues:

### 1\. Blatant Copyright Violations

<iframe width="1236" height="695" src="https://www.youtube.com/embed/6KbqaeAKZqI" title="Kling 2.6 Face Swap with Harley - AI" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

Using AI to recreate a celebrity's likeness without permission is theft of intellectual property, plain and simple.

Imagine if someone used a digital version of Sydney Sweeney to advertise their dodgy cryptocurrency scheme. Sweeney didn't consent. She’s not being paid. Her reputation is being exploited without his knowledge or permission.

A similar case actually happened. A woman in France lost $850,000 after scammers used AI-generated images of [Brad Pitt](https://www.youtube.com/watch?v=ryOKnKCj3as&t=4s) to convince her she was helping the actor.

Actor [Bryan Cranston](https://www.cnbc.com/2025/10/20/open-ai-sora-bryan-cranston-sag-aftra.html#:~:text=OpenAI%20announced%20on%20Monday%20that,Cranston%20said%20in%20a%20statement.) raised concerns after Sora users created deepfakes featuring his likeness without consent or compensation, prompting families of Robin Williams, George Carlin, and Martin Luther King Jr. to also complain to OpenAI.

These aren't just famous people being precious about their image. They're professionals whose likeness has commercial value, and that value is being stolen by anyone with access to deepfake technology.

### 2\. The Primary Use Is Fraud

Strip away the marketing speak and ask yourself: what is this technology actually being used for?

The answer is scams and manipulation.

Since 2017, fraud has accounted for [31%](https://surfshark.com/research/study/deepfake-statistics) of all deepfake incidents. Deepfake fraud attempts spiked by [3,000%](https://keepnetlabs.com/blog/deepfake-statistics-and-trends) in 2023, with losses in North America exceeding $200 million in the first quarter of 2025 alone.

The ability to create convincing videos of famous people saying or doing things they never said or did isn't a feature. It's a weapon. And overwhelmingly, that weapon is being used to harm, deceive, and exploit.

### 3\. The Pornography Problem 

![Screenshot 2026-01-21 at 00.11.57.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/ebf9206ed9050f2f5b70c909e8800c7dc7aa7c5fd0aae357d584698a42c93533.webp)
_(Posted by [@YoRHa2Pee](https://x.com/YoRHa2Pee/status/2013189145816871131) on X)_

Let's address another elephant in the room: the overwhelming majority of deepfake content is pornographic.

The number of deepfake pornography videos produced in 2023 was [464%](https://www.securityhero.io/state-of-deepfakes/#:~:text=Appendix-,KEY%20FINDINGS,07.) higher than in 2022, with almost 4,000 female celebrities found across the top deepfake porn websites.

Analysis shows that [94%](https://www.securityhero.io/state-of-deepfakes/#:~:text=Appendix-,KEY%20FINDINGS,07.) of those featured in deepfake pornography videos are affiliated with the entertainment sector, including singers, actresses, social media influencers, models, and athletes.

When I see people on X casually sharing AI-generated videos of celebrities in suggestive poses or erotic situations, I'm witnessing the normalisation of something deeply unethical: the creation of sexual content featuring real people without their consent.

This isn't parody. It isn't satire. It's digital sexual violence.

If the primary application of your technology is creating non-consensual pornography, you've built something that shouldn't exist.

## The "Just Don't Look" Fallacy
![Photo by Ryoji Iwata on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/ad2f7f5d57d11ab527112cce2b0f42f7921dfd6b3b4cd45fcac33d12e6b67c30.webp)

Some will argue that if you don't like deepfake content, just don't engage with it. Scroll past. Don't create it yourself.

But that ignores the victims. The celebrities whose likenesses are stolen. The ordinary people whose faces end up in pornography they never consented to. The elderly people were scammed out of their life savings by deepfake videos of trusted figures.

Your choice not to create or view deepfakes doesn't protect anyone. It just means you're not personally contributing to the harm.

## What Responsible AI Development Looks Like

I'm not calling for bans. I'm not suggesting we halt all AI progress. That would be hypocritical given my work, and counterproductive given the genuine benefits AI can deliver.

But I am calling for something that seems to be in short supply: **honesty**.

If you're building AI tools, be honest about the likely use cases. Not the best-case scenarios in your pitch deck, but the actual ways people will use your technology in the wild.

If you're deploying AI tools, be honest about whether you've thought through the second and third-order effects. Not just "this is cool," but "this is cool and here's why the benefits outweigh the predictable harms."

And if you're a user experimenting with this technology, be honest about what you're actually doing. Turning yourself into a celebrity for a laugh might seem harmless. 

But you're training yourself and the algorithm that this is normal. That this is acceptable. That consent doesn't matter as long as the technology is fun.

## The Line I'm Drawing
![Photo by Hitesh Choudhary on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/7f353664ab46796727ab0bd96f26fe73b3afb8ddf9519b77db1551f70b8368c7.webp)

I work in AI. I'll continue working in AI. I believe it's one of the most important technologies of our generation.

But not all AI is equal. Not all applications are beneficial. And not every capability that can be built should be built.

Face-mimicking motion AI, as currently deployed, fails every test I can construct for beneficial technology. 

It lacks clear positive use cases. It enables obvious harms. It undermines consent, intellectual property, and our shared ability to trust what we see.

I'm not against AI. I'm against this AI. And I think more people in my industry should have the honesty to say the same.

## Read more
- [I Built Software Wrong for 7 Years. You Shouldn’t Do The Same.](https://videotranslator.ai/blog/i-built-software-wrong-for-7-years-you-shouldnt-do-the-same/)
- [You Sound Like ChatGPT Now](https://videotranslator.ai/blog/you-sound-like-chatgpt-now/)
- [I Rested. Then Shipped An App in 72 Hours. A Story about TrustPRR](https://videotranslator.ai/blog/i-rested-then-shipped-an-app-in-72-hours/)


We crossed a line when faces became interchangeable. 

This AI Should Never Have Been Built

For nearly a decade, I approached every project like an architect designing a skyscraper. 

Grand blueprints. Comprehensive feature lists. Meticulous upfront planning. Everything mapped out before a single line of code was written.

It felt professional. It felt thorough. It felt right.

It was also completely wrong.

After more than 7 years in this field, I've come to realise something that fundamentally changed how I build software: 

**We're not architects. We're gardeners.**

## The Architect's Trap
![Photo by Wiktor Karkocha on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/716aae23434dd3ace952660ff7a82728af007f0050de1857e22f944439c555b5.webp)

The architecture metaphor has dominated software development for decades. We even borrowed the terminology: *software architects*, *system architecture*, *architectural patterns*.

 The influence runs so deep that it shapes how we think about building products.

Here's how the architectural approach typically works: you gather all requirements upfront, design the entire system, create detailed specifications, then hand it over for construction. 

Once the building is complete, the architect's job is done. The structure is fixed, permanent, unchanging.

This made sense when software development was closer to hardware, when programmes were written in machine code for specific mainframe architectures in the 1950s and 1960s. 

Once in production, changes were painful and rare. If the company bought new hardware, you rewrote everything from scratch. That actually resembles building construction.

But ***modern software***? It's nothing like that.

Software is *soft*. Buildings are hard. 

As Andy Hunt and Dave Thomas (2019) observe in The Pragmatic Programmer, rather than construction, software is more organic than concrete. It changes constantly, influenced by user requirements, business needs, market pressures, and technological evolution.

My last project taught me this lesson the expensive way. I approached it like an architect: gathered every possible requirement, defined the architecture upfront, planned for every eventuality. 

![Our previous multilingual video call app, packed with "every feature" (apparently).](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/87a22a83ae7cebb8fcb59b754739980dbd1d91b106923b30a1c224ad0f3a1c39.webp)
_(Our previous multilingual video call app, packed with "every feature" (apparently)._

The result? A big, unwieldy monster plagued by performance issues and technical debt. 

We'd built a skyscraper when we needed a garden.

## The Gardening Mindset
![Photo by Jonathan Kemper on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/33b9a6593514342f74bacbb1805063d174517d18fef1e11821d286f9d0e2e5e9.webp)

So what does it mean to think like a gardener instead of an architect?

**A garden is never finished. It's constantly growing, changing, evolving.** 

You don't plant a garden once and walk away. You tend it continuously, adapting to seasons, weather, and what the plants themselves tell you they need.

The most important aspect of the gardening metaphor is *constant change*. 

Different from a house where an engineer walks away once construction is finished, a garden needs continuous care.

Here's what changed for me when I adopted the gardening mindset:

### 1\. Focus on One Plant at a Time
![Photo by Igor Son on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/36a132fa695d0b68f8c04e6a0552be7f491af5e231f02dda258c60eee136c9a6.webp)

In gardening, you don't try to grow everything simultaneously. You focus on nurturing one plant, making sure it's healthy before expanding to the next.

In software, this means **building one piece of functionality properly**. Not a massive feature set. Not enterprise-grade everything. One thing that works, that users actually need, that you can deploy and observe.

When you deploy this single functionality, it begins to grow and attract attention. Users engage with it. They tell you what they need next. Just like deciding which plants to cultivate based on which ones thrive in your soil and climate.

This is fundamentally different from the architectural approach, where you try to anticipate every need upfront. 

The garden approach lets users guide your growth through real usage, not theoretical requirements.

### 2\. Accept That Things Will Change
![Photo by Y M on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/c9b05bf36acfcfc42ffe60ebc3e25fd85d4575c5920b2fc326e2073b245fcd02.webp)

Buildings resist change. Gardens expect it.

Software is influenced by external factors you cannot control: user requirements shift, business priorities change, frameworks update, security vulnerabilities emerge, competitors launch new features. 

Treating software like a fixed structure means fighting against its fundamental nature.

Gardens change with seasons. Some plants thrive and spread. Others struggle and need to be replaced. You prune to encourage healthy growth. You remove what's no longer working.

In software terms: **you refactor constantly**. You remove features that aren't used. You simplify what's become complex. You rebuild components that have rotted. This isn't failure. It's maintenance. It's how healthy systems evolve.

Research shows that leaving a garden unattended leads to rot, weeds, and eventual unusability. The same applies to software. 

When you don't maintain it, security issues emerge, APIs stop working, and technical debt accumulates until you need to tear everything out and start over.

### 3\. Use the Right Tools for Your Garden
![Photo by Lucas de Moura on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/968a5e571866ffc3af22295567ad8288024b27fa52aff0f08960702d3b3e3f14.webp)

Every gardener has tools they love to work with. These don't have to be the fanciest tools, but they need to work for the type of garden you're cultivating.

You can do everything manually, but certain tools enable efficiency without over-engineering. A gardener doesn't buy industrial farming equipment for a small vegetable patch. They choose tools proportional to the garden's scale.

Software development is the same. You don't need enterprise architecture patterns for a startup MVP. You don't need microservices when a well-structured monolith would do. You don't need every framework and library just because they're popular.

Choose tools that help you tend your garden effectively. Nothing more, nothing less.

### 4\. Understand That Growth Takes Time
![Photo by Sushobhan Badhai on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/16d4044be4160709c8e99b90d35261d90e41187c186604f0e895f62aa343488b.webp)

Architects expect buildings to be completed on schedule. Gardeners understand that plants grow at their own pace, influenced by factors you can control (water, sunlight, soil) and factors you can't (weather, pests, seasons).

Software development works the same way. You can create the right conditions, you can use good practices, but you cannot force growth to happen faster than the system naturally allows.

When stakeholders ask, "How long will this take?" the honest answer is closer to "How long will this tree take to grow if you do everything right?" rather than "How long to build this wall?" 

The first acknowledges environmental factors like legacy code quality, team experience, and unexpected challenges. The second assumes fixed, predictable construction.

This doesn't mean lack of accountability. It means realistic expectations based on organic growth rather than industrial manufacturing.

## What This Means for How You Build
![Photo by Fotis Fotopoulos on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/f58d7da43ddb072acb3343c48b6b86dcabe628434a8b94c7150d7cffa17c59bd.webp)

If you're still thinking like an architect, here's how to start thinking like a gardener:

* **Stop trying to design everything upfront.** You cannot anticipate all requirements. You cannot predict how users will actually behave. Build one thing, deploy it, watch what grows.  
* **Embrace continuous refactoring.** This isn't technical debt or poor planning. It's pruning. It's weeding. It's the essential maintenance that keeps your garden healthy.  
* **Choose simple over comprehensive.** Start with the smallest viable thing that solves a real problem. You can always expand later based on what thrives.  
* **Let users guide your growth.** They'll tell you what they need by how they use what you've built. Listen to the garden, not the blueprint.  
* **Expect change and plan for it.** Don't build rigid structures. Build flexible systems that can adapt as requirements evolve.  
* **Focus on quality over quantity.** One well-tended plant is better than a dozen neglected ones. One polished feature beats ten half-finished ones.

## From My Garden

The product I am building now operates on gardening principles. We started with one piece of functionality. We made it work properly. We watched how users responded.

![Agent dock v1](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/c5659a581f3ebe4eb7579e5eb37f5586af2ba62e8f18b3300991a0e3f1715793.webp)
_(The first iteration of VideoTranslatorAI's app - Instant transcription, translation, and meeting summary.)_

What they told us shaped what we grew next. Sometimes their feedback surprised us. Features we thought were essential turned out to be unnecessary. Capabilities we almost skipped became central to the product.

![Agent dock latest and greatest](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/824e631de03fc3ebe5b68be08c6b1a1cf73f8ceb8e70776b8f6962b6500bcd74.webp)
_(The latest version of VideoTranslatorAI's app - We added summary customisation capability.)_

The codebase stays manageable because we only grow what we need. Performance remains solid because we do not carry the weight of unused features. 

The team stays focused because our priorities emerge from reality, not speculation.

## The Builder’s Choice

You can construct software or cultivate it. Both approaches produce products. But they produce different kinds of products.

Constructed software impresses in demos. It has features for every conceivable use case. It looks complete.

Cultivated software delights in use. It does fewer things, but it does them exceptionally well. It grows with its users rather than guessing at their future needs.

After 7 years of building software, I have stopped trying to be an architect. The blueprints were beautiful. The reality was not.

Now I am a *gardener*. The work is slower. The results are better. And my products actually survive contact with the real world.

Start with one seed. Nurture it properly. Let your users tell you what to grow next.

That is the lesson it took me 7 years to learn. I hope it takes you less.

## Read more
- [You Sound Like ChatGPT Now](https://videotranslator.ai/blog/you-sound-like-chatgpt-now/)
- [I Rested. Then Shipped An App in 72 Hours. A Story about TrustPRR](https://videotranslator.ai/blog/i-rested-then-shipped-an-app-in-72-hours/)
- [A Doctor Told Me Something Haunting](https://videotranslator.ai/blog/a-doctor-told-me-something-haunting/)


the gardening approach to software development

Most engineers think like architects. That’s the mistake. 

I Built Software Wrong for 7 Years. You Shouldn’t Do The Same.

[51 percent](https://english.elpais.com/technology/2024-10-24/humans-are-already-repeating-words-learned-from-chatgpt-such-as-delve-and-meticulous.html).

That's how much more frequently academics started using the word "delve" after ChatGPT launched. Not in their writing, but in their *speech*. On camera. In lectures. In presentations they'd given the same way for years.

"Meticulous." "Adept." "Nuanced." "Realm."

These ChatGPT-favoured words surged across nearly 280,000 YouTube videos from academic institutions. Meanwhile, words the AI uses less, like "bolster," "unearth," and "nuance" as a verb, quietly faded from our vocabulary.

The Max Planck Institute for Human Development documented this. [Researchers](https://www.fastcompany.com/91398460/ai-is-quietly-reshaping-the-way-we-talk) called it the "seep-in effect."

But let's call it what it really is: **reverse training.**

## We are Optimising Ourselves

We like to think we're in control of our tools. We pick them up, use them, put them down. They serve us. We're the ones giving commands.

But here's the uncomfortable truth: 

Every time you rephrase a thought so your AI transcriber catches it cleanly, you've been trained.

Every time you announce "I have three points" before speaking, you've been trained.

Every time you kill a joke because the meeting summariser won't get it, you've been trained.

We've spent years asking how to make AI align with human values. The researchers at Max Planck suggest we've had it backwards: "While intensive research focuses on machines' alignment with human behaviour, our study suggests that the reverse may also be occurring." ([Lopez et al](https://arxiv.org/html/2409.01754v1)., 2024). 

We are aligning ourselves with the machine.

And we're doing it *voluntarily*, *unconsciously*, one optimised sentence at a time.

## How Language Flattening Works

This isn't just about vocabulary. It's about the systematic compression of human expression into machine-readable formats.

UNESCO warns that language flattening is a real phenomenon, where AI-generated content strips away linguistic richness, regional accents, and cultural variations. 

When we optimise our speech for machines, we don't just change our words. We change what we're able to express.

Let me break down how this happens:

### 1\. The Search Engine Effect

![The Search Engine Effect](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/5e7196214412bbaacf9264c2e2cc5ac5d221ffbb06203de5e533fb482b168ab3.webp)

Remember when you learned to search Google? You didn't type "What's the weather like in Sydney today?" You typed "weather London." Two words. No articles. No pleasantries. Pure information retrieval.

You trained yourself to speak Google's language.

Now the same thing is happening with AI tools. You've learned to speak in clear declarations. To announce your intent before expressing it. To avoid the tangential thinking that makes meetings human.

"I want to flag a risk" has replaced "I'm a bit worried about something, and I'm not sure if it's a real issue, but..."

The second version is how people actually think. The first is what an AI transcriber tool captures cleanly.

### 2\. The Loss of Nuance 
![The Loss of Nuance](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/261600bbc4fc5190586a5cb29ae701b0c5b2b987a2648f7b88d87c53b0893def.webp)

When you speak for the machine's benefit, you pay a tax in texture.

Sarcasm becomes risky. The AI might log your joke as a serious statement. ("Meeting notes: Tathagat suggested we abandon the project and go home.")

Cultural idioms become friction. Your "she'll be right" doesn't translate for a globally-trained model.

Emotional subtext becomes invisible. The sentiment analysis sees words, not the pause before them.

[Karelia Vázquez](https://english.elpais.com/technology/2025-11-01/the-chatgpt-effect-weve-all-started-talking-like-robots.html) (2025) put it starkly: "Robotic verbiage erases vulnerability, humour, and everything that makes us human."

What we're left with is communication that's efficient, clear, and utterly sanitised. It's language optimised for machine understanding, not human connection.

### 3\. The Standardisation Trap
![The Standardisation Trap](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/6905d09a98d2fd3bd9087af6f8001d11478a3390e1744d25e6a1eff18b5b7e12.webp)

A [Cornell study](https://news.cornell.edu/stories/2025/04/ai-suggestions-make-writing-more-generic-western) found that Indian English speakers are shifting toward American English patterns when using AI tools. Regional dialects, the linguistic fingerprints of culture and community, are being smoothed away.

This isn't accidental. It's architectural. AI models are trained predominantly on Standard American English. They don't just prefer it. They actively struggle with alternatives.

[The Verge](https://www.theverge.com/openai/686748/chatgpt-linguistic-impact-common-word-usage) documented cases of ChatGPT repeating non-standard English prompts back to users with exaggerated, almost mocking versions of their dialect. One Singaporean user described the AI's response as "super exaggerated Singlish" that was "slightly cringeworthy."

The message is clear: **speak Standard American English, or be misunderstood.**

So people adapt. And the diversity of human expression contracts.

### 4\. The Prompt Engineering of Daily Life
![Prompt Engineer Everyday](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/44b9f4ded6cd93962b63027e00ccc437b7286b3070661c41daf2c537b7880ac2.webp)

Nobody warned us about this, but we're all becoming prompt engineers, not just for ChatGPT or other LLMs. **We're prompt engineering our own thoughts.**

You're in a meeting. You have a complex idea with multiple layers, some uncertainty, and a few contradictory elements. In a purely human conversation, you'd think out loud. You'd explore the idea verbally, circle back, and refine as you go.

But now there's an AI meeting notetaker around you. And you know from experience that if you think out loud, the summary will be a mess. The AI will pick up your half-formed thoughts as if they were conclusions. Your verbal processing will look like indecision in the notes.

So you don't think out loud anymore. You formulate the idea internally first. Then you deliver it clearly, as if reading from a script you just wrote in your head.

You've optimised yourself. But you've also killed the collaborative thinking process that makes meetings valuable.

### 5\. The Empathy Erosion
![The Empathy Erosion](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/2c9c4502d6ee4bbb914d375a1234997eeeb5615453345f71235906d2cdcd4803.webp)

Perhaps most troubling is what happens to emotional intelligence in this new paradigm.

Research from [Cornell University](https://news.cornell.edu/stories/2023/04/study-uncovers-social-cost-using-ai-conversations#:~:text=The%20study's%20lead%20author%2C%20Jess%20Hohenstein%2C%20says,with%20smart%20replies%20were%20evaluated%20more%20negatively) shows that communications suspected of using AI assistance are judged as less cooperative and less affiliative by recipients. 

But here's the paradox: to make AI tools work well, we communicate in ways that trigger exactly this perception.

When you speak in clear, structured sentences with proper grammar and no emotional variance, you sound professional. You also sound like you might be using AI. And increasingly, that makes you sound less trustworthy, less collaborative, less human.

We're trapped in a bizarre loop: optimising our communication for machines makes us seem more like machines to other humans.

## The Cultural Implications

This isn't just about individual behaviour. It's about culture-wide linguistic shift.

The proliferation of AI-generated text is staggering, with vast amounts of content produced daily by algorithms and fed back into systems as training data. 

This creates a recursive loop where AI learns from AI-influenced human speech, which then influences more humans, who produce more AI-influenced speech.

The impact of language models on language flattening may be inversely related to the number of native speakers, meaning smaller languages and cultural groups face even greater pressure to conform to AI-optimised communication patterns.

Think about what we're losing:

* **Regional varieties** that carry centuries of cultural identity  
* **Informal registers** that build social bonds  
* **Creative language play** that drives linguistic innovation  
* **Ambiguity and poetry** that make communication interesting  
* **The mess** that makes us human

In exchange, we get efficiency. Clarity. Machine-readable communication.

Is that really a fair trade?

## A Better Way Forward 
![A Better Way Forward.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/4f31f43e436462b7c9e5646c45b9346a170876d74cb71629ef0113d7c2b33401.webp)

I'm not suggesting we abandon AI tools. I build them myself. Our [meeting agent](https://agent.videotranslator.ai/) transcribes, translates, and summarises conversations in real-time, and it genuinely makes global collaboration easier.

But there's a difference between **using AI as a tool** and **letting AI dictate** how we communicate.

Here's what I've started doing in my own meetings:

**I speak naturally first, optimise never.** If the AI misses something, that's feedback for improving the tool, not a signal that I should change how I talk.

**I flag machine-friendly moments.** When I need something captured perfectly, I'll say "for the record" and then give a clean statement. The rest of the time, I'm human.

**I preserve the mess.** Those tangents, half-formed thoughts, and verbal exploration? They're features, not bugs. If the AI can't handle them, we need better AI, not more robotic humans.

**I check the transcript critically.** When the AI summary misses nuance or context, I add it manually. Yes, it takes time. That time is worth preserving the richness of what was actually said.

The goal isn't perfect transcripts. **It's human communication that happens to be transcribed**.

![Italian-English Agent Interpreter.gif](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/722aaaea7b6fd89c483167339c915daba98848cb5ed5bfa4386d81201d88718f.gif)

## The Script We're Writing

AI is rewriting the human script. Not through force, but through subtle pressure to optimise, clarify, and standardise.

Every time we simplify our language for a voice assistant, we train ourselves to think in simpler patterns. 

Every time we avoid idioms in a transcribed meeting, we lose a bit of cultural expression. 

Every time we speak in "prompts" instead of naturally, we become a little more like the machines we're trying to control.

As AI models become embedded in the architecture of modern life, the very nature of human interaction is being reprogrammed, with our conversations, political debates, and emotional lives being subtly but profoundly reshaped.

The costs of this transformation aren't measured in productivity metrics. They're measured in empathy, human connection, and the shared understanding that underpins society.

Can you see it now?

Listen to yourself in your next meeting. Notice when you're speaking for the humans in the room versus speaking for the AI in the corner. Notice when you catch yourself simplifying, clarifying, standardising.

And then ask yourself: *is this the script you want to follow?*

Because we're writing it, line by line, conversation by conversation. And once it's written, it'll be very hard to unwrite.

The machines are learning from us. But we're learning from them too. And that's a far more profound change than any algorithm.

---
_**Want to use AI meeting tools without letting them flatten your communication?** 
[VideoTranslatorAI](https://agent.videotranslator.ai/) transcribes, translates, and summarises in real-time whilst preserving the nuance of natural conversation. It's built on the principle that AI should adapt to humans, not the other way around. Because the best tools amplify what makes us human rather than training us to sound like machines._

## Read more
- [I Rested. Then Shipped An App in 72 Hours. A Story about TrustPRR](https://videotranslator.ai/blog/i-rested-then-shipped-an-app-in-72-hours/)
- [A Doctor Told Me Something Haunting](https://videotranslator.ai/blog/a-doctor-told-me-something-haunting/)
- [If Translation Is Now Free, Why Pay at All?](https://videotranslator.ai/blog/if-translation-is-now-free-why-pay-at-all/)

AI isn't just changing how we write. It's rewriting how we think, speak, and connect with each other. 

You Sound Like ChatGPT Now 

December 28th. Three days after Christmas.

I was lying on the couch, scrolling through nothing in particular, when I realised something uncomfortable: I was bored.

Not tired. Not burnt out. Just... bored.

For the first time in months, my brain had nothing demanding its attention.

No bugs to fix. No features to ship. No urgent Google Chat messages about my [meeting agent](https://agent.videotranslator.ai). Just empty space.

And in that empty space, an idea appeared.

Three days later, I had a working app.

## The Break Worked (Too Well)
![Took my team to a Christmas lunch just before the holiday!.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/18471b347a69f8a5873f30a1d399cf1b46c6469c757a4c8edd48a6d1ebdf1b79.webp)

A month ago, I wrote [5 Founder-Approved Tips for a Christmas Reset](https://medium.com/@tathagat.banerjee/5-founder-approved-tips-for-a-christmas-reset-2aeabca9e107). 
The core message: **founders need real breaks.** Not "I'll just check email once" breaks.  Real ones.

I took my own advice. 

I switched off. I spent Christmas doing decidedly non-productive things. I went on a long trip with my wife, and it was a lot of fun. 

By December 28th, I was genuinely **recharged**. 

And the funny thing about being recharged? Your brain doesn't just sit quietly. It starts looking for things to do.

## The Idea That Wouldn’t Wait
![Photo by Alicia Christin Gerald on Unsplash .webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/5a5b262ec5c92e7df8a2b3156d8a6081cfb363e5cebb54bee04d22f525915ed8.webp)

I was just scrolling through Twitter (sorry, X, but I'm never calling it that naturally) when I noticed something: everyone was sharing their year-end recaps. 

*Spotify Wrapped* dominated my feed. Since 2016, Spotify's annual recap has become a cultural phenomenon, with users eagerly sharing their listening statistics on social media each December. 

The recap format has been mimicked by rival platforms like Apple Music and Amazon Music, as well as other services like Duolingo and Reddit.

Then it hit me: **developers don't have this.**

Sure, GitHub shows your contribution graph. But it's not shareable in that same satisfying, social-media-friendly way. 

So, I was thinking, how do I turn these commits, pull requests, issues, and code reviews into a social image? Something visual. Something people could actually post.

That's when [TrustPRR](https://trustprr.com/) was born.

![Screenshot 2026-01-14 at 15.56.28.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/c9a3a1ab15bb7ff80b0fa4081f25a2e0297f2a77bde6039aeda5d7621fe7afd0.webp)

## 72 hours to ship

I started building on December 28th. By December 31st, I had a working app.

TrustPRR takes your GitHub activity and transforms it into a polished, shareable image. Your commits. Your PRs. Your reviews. All visualised in a format designed for social media.

![trustprr-share-card (26).webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/03b1214b2f2f80bf54b5e884f21a00438afd173742c7f8668873505e1ea23b1a.webp)

Three days. From idea to deployed product.

Now, I want to be clear about something. This wasn't some superhuman coding sprint fuelled by caffeine and no sleep. It was the opposite.

I was well-rested. My brain was working properly. I wasn't context-switching between twelve different priorities. I had *clarity*.

And that clarity made everything faster.

## Why Rest Enables Speed

Here's what I've learned about building while burnt out versus building while rested.

When you're exhausted, your brain struggles with decisions. Every fork in the road requires conscious deliberation. Should I use this library or that one? Should I structure the code this way or that way? Each choice burns cognitive resources you don't have.

So you stall. You defer. You go down research rabbit holes looking for certainty that doesn't exist. Hours pass without meaningful progress.

A rested brain works differently.

Decisions feel intuitive. You can hold more context in your head at once. You see the whole problem instead of getting lost in pieces. Trade-offs are clearer. The "good enough" threshold is easier to identify.

I've been building things for years. The work itself hasn't changed. But the speed varies wildly based on my mental state.

December 28th through 31st was fast because my mental state was good. Simple as that.

## The ‘Momentum’ Element
![Source: Spotify Newsroom](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/1002399ee3cadd688abd105530fbba6b2c3f154487c75beda3eca782fbd39482.webp)

There's another factor worth mentioning: urgency.

Not manufactured urgency. Not "I *should ship fast because hustle culture says so*." Real, external urgency tied to the idea itself.

Year-end recaps are a cultural moment. Everyone packages their year into something shareable right around now.

That moment has a window. Late December. Early January. Miss it, and the idea loses most of its relevance.

I recognised that window. And I had the capacity to move.

If this idea had struck in October, I might have written it down and returned to it later. But it struck during the window, with a rested brain ready to execute.

Sometimes timing aligns. When it does, you move.

## Where TrustPRR Stands Today

The app exists. You can use it right now at [trustprr.vercel.app](https://trustprr.vercel.app/). 

Don’t worry. **No credit cards needed.** 

You just need to sign in with your GitHub account.

But I'll be honest: it's not polished yet. Three days get you functional, not perfect. I have some UI improvements I'd like to make. Features I want to add. Rough edges I can see clearly now that the build sprint is over.

But it works. It ships. It does the thing.
![What should I add- Need some advice .webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/8d3f16a5444c5fb085d192b7c9ada54692d78889ec839d39465ec022fd631663.webp)
_(What should I add- Need some advice here!)_

I built it primarily to visualise my own GitHub contributions heading into 2025\. The fact that other people might find it useful is a bonus. Maybe it goes somewhere unexpected. Maybe it stays a fun little side project.

Either way, I'm glad it exists\!

## The Meta-Lesson
![Photo by Drew Coffman on Unsplash .webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/06ae76bf7b117ec88ebe62c863a6c3c4f8ec63ac03e6d1bb01663666890fd557.webp)

So what's the takeaway from all this?

**Take your breaks. Real breaks.** 

Step away from your main work completely. Let the compulsive cycle interrupt itself.

But don't feel guilty if ideas find you anyway. Don't force yourself to ignore genuine creative energy just because you're "supposed" to be resting.

The goal is to break the compulsive cycle, not to punish yourself for having ideas.

And if something time-sensitive lands in your lap? If there's a genuine window that won't stay open?

Maybe that's worth three days.

You'd be surprised what a rested brain can build.

Try [TrustPRR](https://trustprr.com) today. It's free. No credit card needed. 

## Read more
- [A Doctor Told Me Something Haunting](https://videotranslator.ai/blog/a-doctor-told-me-something-haunting/)
- [If Translation Is Now Free, Why Pay at All?](https://videotranslator.ai/blog/if-translation-is-now-free-why-pay-at-all/)
- [AI Images Are Everywhere. We’re All Deepfake Suspects Now.](https://videotranslator.ai/blog/ai-images-are-everywhere-were-all-deepfake-suspects-now/)

I Rested. Then Shipped An App in 72 Hours.

Read I Rested. Then Shipped An App in 72 Hours. and explore practical insights from VideoTranslatorAI on multilingual communication, interpretation, and AI-supported workflows.

A doctor in India told me something I still think about.

**"A single mistranslation can end a life."**

He wasn't being dramatic. He was describing his Tuesday. A patient had come in with chest pain. The family translated. Something got lost. The diagnosis came late. The outcome was preventable.

I left that conversation carrying a weight I couldn't shake. Because I realised this wasn't a problem confined to one hospital in one country. It was everywhere. And in Australia, where I now call home, it was hiding in plain sight.

## Not Just A Number

In the 2021 Census, more than [3.4 percent](https://www.abs.gov.au/media-centre/media-releases/2021-census-highlights-increasing-cultural-diversity) of Australians said they don't speak English well or at all. That's not a statistic. That's around 900,000 people who walk into clinics hoping someone will understand them. Hoping their pain translates. Hoping their fear doesn't get lost somewhere between languages.

[Research](https://qualitysafety.bmj.com/content/qhc/33/7/456.full.pdf) tells us that when patients and clinicians can't communicate properly, readmission rates climb. Safety gaps widen. But what the research doesn't capture is the look on someone's face when they realise no one understands what they're trying to say.

I've seen that look. It's not frustration. It's resignation. It's the quiet surrender of someone who has stopped expecting to be heard.

That's not a healthcare system working. That's a healthcare system failing the people who need it most.

## Why This Became Personal![martha-dominguez-de-gouveia-nMyM7fxpokE-unsplash (1).webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/cbfe65ded5411c0058d0f1bb11c126822da5d69f4a3a48bbdbc13fb907d14d79.webp)

I didn't set out to build translation technology. I set out to solve a problem that kept me awake at night.

After those conversations in India, I started paying attention.

I noticed the elderly migrant in aged care, slipping back into her mother tongue as dementia took hold, while staff smiled kindly but understood nothing.

I noticed the young mother in emergency, clutching her sick child, trying to mime symptoms because words had abandoned her.

I noticed families becoming translators by default. Children explaining diagnoses to grandparents. Siblings interpreting medication instructions with Google Translate and hope. The weight of that responsibility, carried by people already scared and exhausted.

This is what language barriers actually look like. Not statistics. But people.

And I decided I couldn't just notice anymore. I had to do something.

## Building Something That Matters

In 2024, something shifted. Through the Small Business Innovation and Research ([SBIR](https://www.nsw.gov.au/grants-and-funding/small-business-research-innovation-sbir-program/recipients)) program, [VideoTranslatorAI](https://videotranslator.ai) began collaborating with South Western Sydney Local Health District (SWSLHD). Their challenge was one we understood deeply: how do you facilitate real communication between clinicians and patients from culturally and linguistically diverse (CALD) backgrounds?![VideoTranslatorAI's SBIR 2022 application](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/3bbfc6f7eb7a3c54d28066d8cafb1a5e390649982e50affd18d143ba7fe399ac.webp)

This partnership validated what we'd believed from the start. The solution couldn't be complicated. It couldn't require training manuals or IT overhauls. It had to work the way conversations work. Naturally. In real time. With dignity intact.

So that's what we built.

**For bedside conversations and aged care rounds**, a clinician speaks English while the patient responds in their own language. Both see subtitles in real time. Both can choose voice translation if they prefer to listen. No one reaches for a phrasebook. No one waits on hold for an interpreter who might not be available.

The patient can describe their pain. Express their preferences. Give informed consent. All in the language that feels like home.

**For telehealth and video consultations**, the same principle applies. We're currently developing this capability because we know that healthcare increasingly happens on screens. A specialist in Sydney should be able to consult with a patient in regional Victoria, regardless of what language that patient grew up speaking.

Each participant selects their language. The conversation flows. Understanding happens.

## What We're Really Building

I'll be honest. We're not building technology for technology's sake. We're building bridges.

The features matter, of course. Dual output with subtitles and voice translation. Custom glossaries for medical terminology that can't afford to be misunderstood. Secure systems that protect sensitive conversations.

But what matters more is what these features make possible.

An elderly patient asking for comfort in Mandarin, and being understood.

A parent grasping their child's treatment plan in Arabic, without the terror of missing something critical.

A clinician finally able to ask, "Where does it hurt?" and receive an answer they can trust.

This is what inclusive healthcare looks like. Not as an aspiration, but as a practice.![Italian-English Agent Interpreter.gif](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/722aaaea7b6fd89c483167339c915daba98848cb5ed5bfa4386d81201d88718f.gif)

## The Future We're Working Toward

Australia's diversity isn't slowing down. One in four Australians speaks a language other than English at home. Aged care facilities are filling with migrants who built this country and now deserve to be cared for in languages they understand.

The question isn't whether we need multilingual healthcare. The question is whether we're willing to build it.

I think about that doctor in India often. The weariness in his voice when he described what mistranslation costs. The way he said "preventable" like a word that had lost its meaning.

We can't fix everything. But we can fix this.

Every conversation that happens in someone's own language is a small victory against a system that too often leaves people behind. Every diagnosis that lands clearly, every consent form that makes sense, every moment of genuine understanding between caregiver and patient.

These moments add up. They become trust. They become better outcomes. They become the kind of healthcare that doesn't ask people to leave their identity at the door.

That's what we're building at VideoTranslatorAI. Not just translation. Connection.

Because in medicine, words carry weight. And no one should have to carry that weight alone.

*If you're working in healthcare or aged care and want to explore how real-time translation could support your patients, I'd love to talk.If you have questions, [try our product now](https://agent.videotranslator.ai/chat).*

## Read more
- [VideoTranslatorAI Advances to the Proof of Concept Phase in the NSW SBIR Program!](https://videotranslator.ai/blog/videotranslatorai-advances-to-the-proof-of-concept-phase-in-the-nsw-sbir-program/)
- [AI Images Are Everywhere. We’re All Deepfake Suspects Now.](https://videotranslator.ai/blog/ai-images-are-everywhere-were-all-deepfake-suspects-now/)
- [I'm Done Pivoting. Here's What I'm Building in 2026.](https://videotranslator.ai/blog/im-done-pivoting-heres-what-im-building-in-2026/)

Lost words. Lost lives. Silent crisis in Australian Healthcare

And it leads me to build a technology that listens  

A Doctor Told Me Something Haunting

Have you noticed? When you scroll through X (formerly Twitter), Instagram, or Facebook these days, nearly every post in a foreign language comes with a convenient "Translate" button. Click it, and boom, instant understanding.

X expanded [Grok translation](https://www.socialmediatoday.com/news/x-formerly-twitter-expands-grok-translations-posts/757620/) to all US users in August 2025, automatically translating posts from languages around the world into English. [Meta](https://www.techradar.com/computing/artificial-intelligence/meta-announces-an-ai-translation-tool-that-could-change-the-way-you-watch-instagram-and-facebook-reels-forever) rolled out AI translation tools that automatically translate audio in Reels, complete with voice simulation and lip syncing, now supporting English, Spanish, Portuguese, and Hindi.

For anyone who remembers struggling to communicate across language barriers, this feels like magic. 

A genuine, measurable improvement in how humans connect with each other. The language wall that separated billions of people is crumbling, one auto-generated caption at a time.

I should be celebrating. Instead, ***I am rethinking my entire business.***

## The Awkward Position of Building What Becomes Free
![Our early video translator app (now deprecated)](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/dc6fe84630040709a601f320467d54bbf947101de8c00393e0944a4d9f632f0f.webp)

I build AI translation tools for a living. [VideoTranslator.ai](https://videotranslator.ai/) is my company. And the technology I have spent years developing is now being given away by platforms with deeper pockets and larger user bases.

This is not a complaint. It is just reality. When Instagram can translate every story, and YouTube can caption every video, the market for basic translation shrinks. Dramatically.

This realisation forced builders like me to ask harder questions. 

***If basic translation is commoditised, what value can we actually provide?***   
***What can dedicated translation tools do that a social media platform's built-in feature cannot?***

The answer, I have discovered, lies in moving beyond translation as a product and reimagining it as a capability. 

A foundation for things that were previously impossible

## 7 Reasons AI Translation is Still Worth Paying For

We all know that free things always come with limitations. Relying solely on these free options means you’re likely missing the essential, advanced feature that only a paid service can provide. 

Here are seven AI translation capabilities you might want to invest in. 

### 1\. Enhance Cross-Cultural Communication
![Photo by Campaign Creators on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/eb6aab2fe77aca18e03c7a0fdcd4d07d81d52152845eccce428e73b5494adeec.webp)

Platform translations handle surface-level meaning, but they completely miss the subtleties that make or break international business relationships.

Modern translation tools now offer real-time voice translation with tone preservation. Imagine negotiating a deal with a Japanese client. The platform translation gives you the words, but it strips away the politeness levels that are crucial in Japanese business culture. 

A dedicated tool? It maintains the formal register, preserves honorifics, and flags when your casual phrasing might come across as disrespectful.

Tools like [JotMe](https://www.jotme.io/) offer contextual translation engines that read multiple sentences as a cohesive unit to preserve tone, flow, and meaning, especially useful in professional or emotionally nuanced communication. 

This isn’t just helpful. It’s the difference between closing a deal and accidentally insulting your potential partner.

### 2\. Boost Learning and Skill-Building
![A conversation with Tutor Lily app (Source: BBC News)](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/404bb8eb69f28a1cfa395c21fa3b67509fd76d9526ea5e7fc67aa0d2002bb2e3.webp)

Here's something social media translation will never do: help you actually learn the language.

Advanced translation tools have become educational platforms. You can translate your own writing back and forth, comparing versions to understand how grammar structures differ between languages. The tool explains why certain idioms don't translate literally and suggests culturally appropriate alternatives.

Want to learn Spanish? Write an email in English, translate it, then translate it back. The discrepancies reveal exactly where your understanding gaps are. 

Some tools now include grammar explanations and idiomatic breakdowns, turning every translation into a learning opportunity.

This transforms translation from a crutch into a stepping stone.

### 3\. Create and Adapt Content for Multiple Markets
![Photo by Austin Distel on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/b1dd846c0b5eedac1965596397afb22344d0505fda61905ff4986bddfd36cea7.webp)

Social media platforms translate for consumption. Professional tools translate for creation.

If you're a content creator or marketer, you need more than word-for-word conversion. You need localisation—adapting humour, references, and messaging for different cultural contexts. A joke that kills in the US market might fall flat in Germany or offend in Japan.

Brands that rely on AI translation alone often end up with content that sounds awkward, off-putting, or even offensive in the target language. Professional tools let you generate multilingual versions of your social posts, blogs, or marketing materials, then tweak them for cultural relevance. 

You're not just changing languages. You're adapting the entire message.

### 4\. Maintain Brand Voice Across Languages
![Photo by UX Indonesia on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/9b6afee601272e68dc39ff3bf0566a1c7534d4be21c2d165e8a8c109ed69b47f.webp)

Here's something free translation misses entirely: **consistency**.

If you're running a global brand, you need every piece of content to sound like you, regardless of language. That means specific terminology, a consistent tone, and industry jargon that's translated correctly every single time.

Modern translation tools offer custom glossaries and translation memory systems. Once you establish how certain terms should be translated, the tool remembers. Your "customer success team" doesn't randomly become "client happiness squad" in French because some algorithm thought it sounded friendlier.

As users make more edits over time, they increasingly train their machine translation engine, resulting in more expansive translation memories that auto-populate for future translations. This learning capability is something social media platforms will never offer individual users.

### 5\. Access Professional Documentation Translation
![Photo by NordWood Themes on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/41ff8f57b4d47986f5820bbad46bacc1bfc46b866498c248e38e21187ff07976.webp)

Try uploading a 50-page legal contract to a free tool for translation. Go on, I'll wait.

Right. That's not happening.

Professional translation tools handle complex documents (legal contracts, medical records, technical manuals) with formatting intact. They maintain tables, preserve footnotes, and keep your carefully crafted document structure instead of spitting out a wall of unformatted text.

AI translation models can make educational materials and lectures available across the globe, whilst allowing companies to provide multilingual customer support. The infrastructure requirements for this level of translation are significant, something free social platforms aren't equipped to provide.

### 6\. Enable Real-Time Collaboration Across Languages
![Photo by Chris Montgomery on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/3a0bfac0735eb4bcba71d13a8310545b23903925da065b39ea50adc8380024f6.webp)

Picture this: you're in a virtual meeting with team members from five different countries. Everyone speaks their native language, but everyone understands each other perfectly in real-time.

Advanced tools transcribe and translate meetings in real time, displaying both spoken content and translations side by side, with all transcripts saved for easy access. Platform translations can't do this. They're designed for asynchronous text consumption, not live, multi-party conversations.

Professional tools integrate with Zoom, Teams, and Google Meet, providing live interpretation, meeting summaries, and searchable transcripts. They turn language barriers into minor inconveniences rather than deal-breakers.

### 7\. Protect Privacy for Sensitive Communications
![Photo by Sue Winston on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/195f73ad6a1fbabead12facfea80d6ae462623f47bf0a9f578ae615ae96abe2a.webp)

Here's an uncomfortable question: do you really want to paste confidential business information into a free translation platform?

Every time you click "translate" on a platform, that content passes through their servers. For casual content, that's fine. For sensitive business negotiations, medical information, or legal matters? Not so much.

Professional translation tools offer enhanced security measures, private servers, and compliance with data protection regulations. They're not feeding your confidential information into algorithms that improve their general translation models—or worse, training data for other features.

## A Tool That Grew With the Challenge

At [VideoTranslator.ai](https://videotranslator.ai/), we have been living through this evolution firsthand. When basic translation became ubiquitous, we asked ourselves what people actually needed help with.

The answer, increasingly, was **meetings**.

Real-time conversation remains one of the hardest translation challenges. The speed, the interruptions, the need to capture not just words but action items, decisions, and commitments. Platform translation features handle pre-recorded content well, but live conversation is different.

So we built an **[AI chat agent](https://agent.videotranslator.ai/login)** designed specifically for this use case. It ***transcribes*** in real time across languages, ***translates*** on the fly, and, perhaps most valuably, ***summarises*** the outcomes. Who agreed to what. What decisions were made. What needs follow-up.

![IMeeting assistant tool by VideoTranslatorAI
](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/722aaaea7b6fd89c483167339c915daba98848cb5ed5bfa4386d81201d88718f.gif)

Does it replace human interpreters for high-stakes diplomatic meetings? No.   
Does it handle every possible edge case perfectly? Also no. 

But for the day-to-day reality of working across language barriers, whether it’s a quick check-in with your Spanish-speaking supplier or a product demo for a potential client in Japan, it removes enough friction to make those conversations actually happen rather than being postponed or avoided.

The goal wasn't to create the most sophisticated translation engine in the world. It was to build something that addresses the gap between "good enough for casual browsing" and "hire a professional translator for thousands of dollars." Most business communication lives in that middle ground.

This is what the evolution of AI translation looks like in practice.   
Moving from tool to teammate.

## The Language Wall Is Down. Now We Build.
<iframe width="1236" height="695" src="https://www.youtube.com/embed/zqqsg52rnnI" title="A Big Reveal: What the Agentic Version of VideoTranslatorAI Looks Like!" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

The original promise of AI translation was breaking down language barriers. That promise has largely been fulfilled. Translation is everywhere, often invisible, increasingly free.

But breaking down walls was not the end goal. It was the beginning.

What matters now is what we build in the space where those walls used to stand. Deeper cross-cultural understanding. Genuine global collaboration. Access to the full breadth of human knowledge and creativity.

If in 2026 you’re still using translation tools the way you did five years ago, you’re missing what they have become. Not just converters, but capabilities. Not just products, but platforms for things that were previously impossible.

Basic translation is dead. And what comes after is worth paying attention to.

## Read more
- [AI Images Are Everywhere. We’re All Deepfake Suspects Now.](https://videotranslator.ai/blog/ai-images-are-everywhere-were-all-deepfake-suspects-now/)
- [I'm Done Pivoting. Here's What I'm Building in 2026.](https://videotranslator.ai/blog/im-done-pivoting-heres-what-im-building-in-2026/)
- [It was Supposed to Happen in 2027. It Arrived Early.](https://videotranslator.ai/blog/it-was-supposed-to-happen-in-2027-it-arrived-early./)


Everyone is using better AI translation now

Social media made translation free. Here's why paid tools still matter in 2026. 

If Translation Is Now Free, Why Pay at All?

Scroll through Twitter (or X, if you're pedantic) for five minutes, and you'll see AI-generated images created by Grok everywhere. 

Browse through your Instagram feed and count how many images were touched by Midjourney or DALL-E. 

Check your LinkedIn connections' profile pictures. Spot the telltale smoothness of AI enhancement.

The numbers are staggering: approximately [34 million](https://www.digitalsilk.com/digital-trends/ai-statistics/) new AI-generated images are created every day. The quality has improved exponentially. Gone are the days of obviously distorted hands and uncanny valley faces. 

Modern AI image generators produce photos so realistic that even AI-savvy adults can only identify AI-generated images about half the time. The technology has become so sophisticated that [71%](https://www.forbes.com/sites/bernardmarr/2025/03/10/15-mind-blowing-ai-statistics-everyone-must-know-about-now/) of images shared on social media globally are now AI-generated.

This isn't some distant future scenario. It's happening right now, and it's accelerating. With this explosion came the inevitable response: **AI image detectors,** positioned as the solution to our authenticity crisis. 

Tech companies rushed to market with tools promising to separate the real from the synthetic. Investors poured money into detection startups. Platforms began implementing automated checks.

*We're about to make the same mistakes all over again.*

## The AI Writing Detector Disaster We're Ignoring
![Some of the most popular AI writing detectors.](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/1bf63caba2b4cb2b2d9ed6a07754ea070462673f82bad829a9dcbcaa72457d0f.webp)

Remember when AI writing detectors were going to save academic integrity? Universities rushed to implement tools like GPTZero and Turnitin, confident they could catch students using ChatGPT. 

In October 2025, [ABC News reported](https://www.abc.net.au/news/2025-10-09/artificial-intelligence-cheating-australian-catholic-university/105863524) that Australian Catholic University had used AI detection technology to accuse approximately 6,000 students of academic misconduct in a single year. The problem? **Many of those accusations were wrong.**

A paramedic student found himself fighting to prove his innocence after being flagged for cheating on work he'd written himself. The burden of proof had shifted. Suddenly, students weren't presumed innocent until proven guilty. They were presumed guilty until they could somehow prove they hadn't used AI.

This isn't an isolated case. [NPR reported in December 2025](https://www.npr.org/2025/12/16/nx-s1-5492397/ai-schools-teachers-students) that school districts across the United States are spending thousands of dollars on AI detection tools *"despite research showing the technology is far from reliable."*

At the University of Houston-Downtown, a computer science student named Leigh Burrell received a zero on an assignment worth 15% of her grade. Her professor believed she'd used ChatGPT.  But [The New York Times](https://www.nytimes.com/2025/05/17/style/ai-chatgpt-turnitin-students-cheating.html) reviewed her Google Docs editing history and found she'd drafted and revised the paper over two days. She'd done the work. The detector got it wrong anyway.

[Non-native English](https://arxiv.org/pdf/2304.02819) writers. Neurodivergent students. Anyone whose writing style happened to align with statistical patterns the detectors were trained to recognise. They all became suspects. 

[OpenAI](https://openai.com/index/new-ai-classifier-for-indicating-ai-written-text/) itself quietly retired its AI text detection classifier after achieving just a 26% success rate. Think about that: 

**The company that created ChatGPT couldn't reliably detect its own AI's output.**

Yet here we are, ready to deploy image detectors with the same blind faith.

## Why AI Image Detectors Will Follow the Same Path
![Photo by Nahrizul Kadri on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/a19a552586908ebf3b22314769dcb7b082ec0358298a71398a3a991512bd79b7.webp)

The fundamental problem is identical: AI image generators are constantly improving, and as these platforms evolve, they create increasingly difficult images for detectors to flag. It's an arms race where the detection tools are perpetually playing catch-up.

Current AI image detectors perform no better than a coin toss in many cases. Researchers at Cornell University developed the [Visual Counter Turing Test](https://arxiv.org/abs/2411.16754), a benchmark testing 17 leading AI image detection models against 166,000 images.

The result? An average accuracy of just 58%. That's 8% points better than random guessing.

Even more troubling: a study from [Nexcess](https://elearn.eb.com/real-vs-ai-images/) found that even AI-savvy adults could identify AI-generated images only about half the time. 

If humans who understand AI struggle to distinguish real from synthetic, why are we confident that automated tools will do better?

## The Paranoia Economy and Weaponisation of Doubt

Here are two things that worry me most.

First, the emergence of a ***paranoia economy*** where everyone becomes suspect.

Imagine you're a **photographer** who spent hours capturing the perfect wildlife shot. You submit it to a competition. The organisers run it through an AI detector. The algorithm flags it. Suddenly, you're defending your credibility, explaining your workflow, providing metadata, and essentially proving your innocence.

Or perhaps you're a **graphic designer** creating marketing materials. A client runs your work through a detector. False positive. Now you're justifying your artistic choices and professional expertise, all because an algorithm thinks your colour grading looks "too perfect" or your composition is "statistically improbable".

This isn't hypothetical fearmongering. We've already seen it happen with text detectors, where students had to prove they wrote their own essays. 

![Photo by Levi Arnold on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/e729844cb93ea6b8b90fe41a69a25ef435ca5a125646e307304fca04bd8c6078.webp)

Second, the ***weaponisation of doubt.*** 

In August 2024, Donald Trump looked at a photograph of thousands of people gathered at a Kamala Harris rally in Detroit. “There was nobody at the plane, and she ‘A.I.’d’ it,” he wrote on Truth Social. The crowd, he claimed, “DIDN’T EXIST!”

![Trump's original post on his Truth Social account](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/cb624bc989ada79c69df8b1ccde4497b5c26ca3fd46979b85563e4617684d91a.webp)

[BBC Verify confirmed](https://www.bbc.com/news/articles/cx2lmm2wwlyo) the rally was real. The [New York Times](https://www.nytimes.com/2024/08/11/us/politics/trump-harris-crowds-ai.html) confirmed it. Dozens of attendees posted their own photos and videos from inside the crowd. News crews filmed the entire event.

None of that mattered. The accusation spread and sparked debate on social media. 

An [AI researcher](https://www.npr.org/2024/08/14/nx-s1-5072687/trump-harris-walz-election-rally-ai-fakes) told NPR: 

***"This is a photo of an event in one city on one day. I mean, what hope do we have to actually tackle complex problems in society if we can't agree on this?"***

This is the endgame of unreliable detection. Not a world where we catch every fake, but a world where nothing can be proven real. Where the existence of AI images creates enough uncertainty to dismiss anything inconvenient.

## When AI Image Detectors Actually Help

I'm not arguing that these tools have no place in our digital ecosystem. They do. 

There are cases where it serves the public interest. But those cases share specific characteristics worth understanding

In May 2023, an image showing black smoke billowing from what appeared to be an explosion near the Pentagon went viral. Verified Twitter accounts shared it. A fake Bloomberg news account amplified it. The stock market dipped before authorities confirmed no explosion had occurred.

![Pentagon in (Fake) Flames. Source: Bloomberg](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/d2b04ef22b703983bf6622eda307a4773f2f881ff8a9cc83117e1b7f793d243e.webp)

Experts quickly identified it as AI-generated. The fence looked warped. The smoke seemed unnatural. And crucially, no corroborating footage existed. No one else photographed the explosion because it never happened.

Two months earlier, an image of Pope Francis in a stylish white Balenciaga puffer jacket fooled millions before being identified as a Midjourney creation. Again, the giveaway wasn't just visual inconsistencies. It was the absence of any other photograph from any other angle.

![A viral AI-generated picture of the Pope wearing a puffer jacket. (Source: X).webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/0f133ca39f4e5f0c059847bcfdafe2458bf092b9a7b75ac59b1aede792023c0a.webp)

**These cases show when detection helps:**

* Extraordinary claims (explosions, celebrities behaving unusually, breaking news) with no corroborating evidence.  
* Single images from unverified or suspicious sources.  
* Situations where absence of other documentation is itself suspicious

**Detection fails when:**

* Multiple independent sources document the same event (like the Harris rally).  
* Verification teams have already confirmed authenticity  
* The detection result contradicts abundant real-world evidence

Never use detection to dismiss well-documented events. In that context, citing AI detection tools to claim the event was fake isn't scepticism. **It's denial.**

## How to Evaluate Suspicious Images Yourself
![Photo by Geri Mis on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/422b724e92f8fc57d157bea0ad2e302ddd2c44d10846bd7d30e09a412821c05b.webp)

Rather than relying solely on automated detectors, here's a more practical approach:

**1\. Reverse image search.** Upload the image to Google Images or TinEye. If it's a real photograph, you'll often find the original source, the photographer's credit, or news coverage that includes the same image.

**2\. Check for multiple angles.** Real events are photographed by multiple people. The Harris rally Trump called fake? Dozens of attendees posted their own photos and videos from different positions in the crowd. A single suspicious image with no corroborating footage deserves scepticism.

**3\. Look for source verification.** Major news organisations like the BBC, Reuters, and AFP have verification teams. If they've confirmed an image is authentic, that carries more weight than an AI detector's probability score.

**4\. Examine the context.** Who shared this image first? What's their motivation? An image from a verified journalist at a news outlet has different credibility than an anonymous post on social media.

**5\. Check obvious AI tells.** While AI-generated images are improving rapidly, many still have issues with hands (wrong number of fingers, awkward positioning), text (garbled or nonsensical words), and background details (objects that melt into each other, inconsistent shadows).

These methods aren't foolproof. But they're more reliable than trusting a tool that's wrong 42% of the time.

## The Solution: Voluntary Transparency
![AI labeling on Facebook. Source: Meta](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/3bc1eda4f25dab57f2d59e8bda93a6b788e1d8ecc26eb4f54d11775995e43268.webp)

Rather than creating a surveillance state where every image is suspect until proven innocent, we need a cultural shift towards voluntary transparency.

Imagine if platforms made it easy for creators to flag their own AI-generated work. Not as a scarlet letter, but as a simple informational label. "This image was created with AI assistance." "This photo was enhanced using AI tools." "This is a completely AI-generated image."

Some platforms are already moving in this direction. Meta has begun labelling AI-generated images on its platforms, and many AI tools now embed invisible watermarks. But this only works if it becomes normalised rather than stigmatised.

The key word here is "**voluntary**". When people feel they'll be punished for using AI tools, they hide their usage. When we create an environment where AI assistance is accepted and expected to be disclosed, transparency becomes the norm.

![AI labelling on Instagram. Source: Meta](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/6079d8852b8faa76414b61a4aca0c4af1115668b7d4ad9505a0aaf9934207dea.webp)

## Where We Go From Here

The parallel with AI writing detectors isn't coincidental. Both technologies emerged from the same impulse: the desire for certainty in an increasingly uncertain digital landscape. We want to know what's real, what's human, what's authentic.

But certainty is precisely what these detectors can't provide. The detector's model is only as good as the database it was trained on. 

As AI generators evolve, detection models must constantly update to keep pace. There will always be a lag, always be errors, always be people caught in the crossfire.

We need to accept that we're entering an era where perfect detection isn't possible. What matters isn't whether an image was technically created by AI, but whether it's being used honestly. 

A clearly labelled AI-generated illustration for a blog post isn't misinformation. An AI-generated photograph passed off as documentary evidence is.

Context and intent matter more than the technical means of creation.

## A Final Thought

I'm not against AI image detectors existing. They have their place in the professional toolboxes of fact-checkers, journalists, and platform moderators. But the wholesale deployment of these tools as gatekeepers, with the power to brand people as frauds based on probabilistic assessments, will create more harm than good.

We watched it happen with writing. Students falsely accused. Careers damaged. Trust eroded.

Let's not make the same mistake with images.

The technology isn't the problem. Our blind faith in it is.

## Read more
- [I'm Done Pivoting. Here's What I'm Building in 2026.](https://videotranslator.ai/blog/im-done-pivoting-heres-what-im-building-in-2026/)
- [5 Founder-Approved Tips for a Christmas Reset](https://videotranslator.ai/blog/5-founder-approved-tips-for-a-christmas-reset/)
- [It was Supposed to Happen in 2027. It Arrived Early.](https://videotranslator.ai/blog/it-was-supposed-to-happen-in-2027-it-arrived-early./)


Anyone can now call your photos ai. Here is why

Why AI image detectors are repeating the same disastrous mistakes as text detectors 

AI Images Are Everywhere. We’re All Deepfake Suspects Now.

If I'm honest, 2025 didn't go the way I planned. And somehow, that's exactly why I'm grateful for it.

This year brought more ups and downs than I could have predicted. We pivoted our product. We celebrated wins that felt hard-earned. We learned lessons that stung at first but now feel like gifts. 

There were moments I questioned everything, and moments I couldn't believe how far we'd come.

So as I sit here wrapping up the year, I'm not just relieved. I'm genuinely joyful. 

The ups and downs didn't break us. They shaped us. And now I'm looking at 2026 with the kind of energy I haven't felt in a while.

## Putting My Vision on Paper

![Vision board 2026](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/910f4661799ae43393bb6a09fcf910a5ebd27aa15dd0a9b2b0c0ea6eb2de5304.webp)
*My 2026 vision board, big thanks to Canva’s template.* 

Here's what I've learned: As founders, we live in a state of **controlled chaos**. Our to-do lists multiply. Our priorities shift weekly. Our brains juggle a hundred things at once.

That's exactly why I started using **vision boards.**

Not the fluffy, magazine-cutout kind (but hey, no judgment if that’s your jam\! If I had a flair for aesthetics, I’d dive right into that world too). 

What I mean is a clear, visual reminder of where I'm headed. Something that cuts through the noise when everything feels overwhelming.

Vision boards work because they force clarity. They make you commit to what actually matters. And when you're deep in founder mode, that clarity becomes your compass.

So here's what I'm calling in for 2026\.

## App & Product Growth: Building What People Actually Use
![App and Product Growth vision](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/6adbdffcac75bd0b6a58c985cb7e8a1ffb9b3aa933f2adc6a57e87864e0af395.webp)

**The Vision:**

* Launch our multi-agent AI translation platform on both web app and mobile app  
* Integrate video call capability with real-time translation  
* Hit our first 100 users who actually stay (not just sign up and ghost)

This feels simultaneously exciting and terrifying. We've spent 2025 learning what users need. Now it's time to build it properly and get it into hands that will actually use it.

**How I'm Making This Happen:**

**Continue the grind, but smarter**. I've learned that grinding without direction is just expensive procrastination. 

In 2026, I'm committing to focused building sprints with clear milestones. Two-week cycles: build, ship, gather feedback, iterate. No more "we'll launch when it's perfect" because perfect never arrives.

**Ask for feedback ruthlessly.** This year taught me that users will be brutally honest if you ask the right questions. 

Instead of "*do you like this feature?*", I'm asking "*when was the last time you needed real-time translation and what frustrated you about existing solutions?*" The answers guide better building.

**Ship productively, not just frequently.** Shipping fast is valuable. Shipping things users actually need is more valuable. 

I'm setting up a feedback loop where we're talking to potential users weekly, building based on patterns we see, and validating before we scale.

The mobile app launch is ambitious, but we're not starting from scratch. We've learned what works on web. Mobile just requires rethinking the interface for different use cases: on-the-go translation, quick video calls, offline capability for travellers.

Those first 100 staying users? That's the real target. Sign-ups are vanity metrics. Active users who come back, who recommend us to others, who actually use the product in their workflow? That's validation worth celebrating.

## Community & Impact: Building in Public and Helping Others
![Community and Impact vision](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/e9fa9b92a4a2ac3fb977878b32c4cb4fa49710965da9326825c098bf091ee1e9.webp)

**The Vision:**

* Build in public more actively, especially on X (Twitter)  
* Share founder insights and lessons consistently on Medium  
* Help other founders by creating content that's actually useful, not just inspirational fluff

I started building my X and Medium presence in Q4 2025, and honestly, I wish I'd started earlier. The conversations I've had, the connections I've made, the feedback I've received \- these platforms aren't just marketing channels. They're communities where founders help each other survive and thrive.

**How I'm Making This Happen:**

Be more active on X with concrete goals. I'm committing to sending more valuable replies and posting daily. 

Things I will post? insight from building, lessons learned, questions to the community. No generic motivational quotes. Real experiences, even the messy ones.

![My twitter account’s performance (1 October- 18 December 2025)](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/d9b6f0507009152bbb051ad1b108c14e74ad6dfcb583723f04de343d5ed307fd.webp)
_My twitter account’s performance (1 October- 18 December 2025)_

My Q4 stats showed something I should have known earlier: growth isn't about posting perfect content. It's about showing up in conversations. I became the "*reply guy*" (commenting on other founders' posts, asking questions, adding value to discussions). Engagement grew because I engaged first. 

So 2026 is about continuing that: showing up, contributing to conversations, being present in the community rather than just broadcasting into the void.

Medium will be my long-form platform for deeper dives. I'm aiming for at least two substantial articles weekly: one about AI translation developments, one about founder lessons.

![Medium statistic - November 2025](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/e8f9e765a887e6a4575f35f752dabe8c1484f0a687756a9a1efad292b810f943.webp)
_My Medium performance in November 2025. My best performance, so far!_

Building community isn't about accumulating followers. It's about creating value that helps others. 

When I share how we handled a product pivot, I'm hoping it saves another founder the time or mistakes we made.

The best part? **Building in public forces accountability.** 

When you publicly commit to goals, you're more likely to actually do the work. And when you share struggles, people offer solutions you hadn't considered.

## Wild Card/Fun: Speaking, Summits, and Big Swings
![Wild Card and Fun Vision](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/b18c1f78853d0576d7f47fb79ffc1562712eed60aec9b7f0b5ca122ab7c0020c.webp)

**The Vision:**

* Speak at a conference about AI translation, founder lessons, or building in public  
* Host or participate in more community events like the Women in Sensing Summit  
* Land another major project or grant that pushes our capabilities

Not everything worth chasing shows up on a product roadmap. 2025 gave me some of my most memorable moments outside of building. 

We were absolutely delighted to host the networking reception of the [NSSN Women in Sensing Summit 2025](https://www.nssn.org.au/upcoming-events/2025/5/15/women-in-sensing-summit-leaders-in-sensing-frontiers). And attending the NSW Commercialisation Showcase 2025 filled me with fresh ideas and new connections.

\[[https://www.sbs.com.au/news/small-business-secrets/article/as-fire-season-looms-not-everyone-gets-the-message-a-new-project-aims-to-change-that/pr7rz9to7](https://www.sbs.com.au/news/small-business-secrets/article/as-fire-season-looms-not-everyone-gets-the-message-a-new-project-aims-to-change-that/pr7rz9to7)\] 

We even got featured on SBS Australia News in 2024 thanks to the NHTP programme, which was surreal and validating in ways I didn't expect. Those experiences taught me that the "wild card" goals often create the most meaningful growth.

I'm forever grateful for the opportunities we had in 2024-2025, like being part of NSW [Small Business Innovation & Research (SBIR)](https://www.nsw.gov.au/grants-and-funding/small-business-research-innovation-sbir-program/recipients) and [Natural Hazards Technology Program (NHTP)](https://www.chiefscientist.nsw.gov.au/industry-development/innovation-procurement/natural-hazards-technology-program). Those programs changed our trajectory.

For 2026, I want to keep that energy going.

**How I'm Making This Happen:**

Actively monitor NSW grant opportunities and events throughout 2026\. I've created alerts for keywords: translation, AI, language technology, multicultural services, accessibility. When relevant opportunities appear, I'm applying early with thoughtful proposals that show we understand the problem deeply.

The Women in Sensing Summit 2025 was incredible because we created space for conversations that weren't happening elsewhere. In 2026, I want to either host another event or partner with organisations doing similar community-building work. There's something powerful about bringing people together in person, even in our digital-first world.

I'm also reaching out to conference organisers proactively. Not just "I'd love to speak" emails, but specific proposals: "I can share lessons from building AI translation products for multicultural communities" or "Here's what we learned hosting technical summits that actually engage attendees."

The key difference between dreaming and doing is taking the first step. For speaking, that's reaching out to three conference organisers per month. For grants, that's applying to every relevant opportunity with strong proposals. For community events, that's having conversations now about what might be possible later in the year.

## What I'm Calling In for 2026
![Photo by Tim Mossholder on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/0acc6d952b1479caf2c3bcfc0e3398da2dc6a7135bbfe2ff0616a376a5a4a312.webp)

I'm calling in a year of focused building, community connection, and strategic risk-taking.

I'm calling in those 100 staying users who actually use our multi-agent translation platform, who come back daily, who tell other people about it because it genuinely solves their problem.

I'm calling in meaningful conversations on X and Medium that help other founders avoid mistakes I made, or encourage them to take swings they're hesitant about.

I'm calling in a conference stage where I can share what we've learned building AI translation products, and maybe inspire someone else to start building what they're dreaming about.

I'm calling in another major project like SBIR or NHTP that pushes our capabilities beyond what we thought possible, that forces us to grow in uncomfortable but valuable ways.

Most importantly, I'm calling in the **discipline** to actually do the work required to make these things happen. Vision without execution is just daydreaming. Execution without vision is just busywork. I need both.

## Now It's Your Turn

This is what I'm building toward in 2026\. What about you?

Whether you're a founder, a builder, someone pivoting careers, or just trying to grow in intentional ways, I'd love to hear: **what's your top goal for 2026, and what's one concrete action you're taking this week to move toward it?**

Drop it in the comments. Let's make this year about collective building, not just individual grind.

Here's to a year of building things that matter, connecting with people who get it, and taking swings that feel impossible until they're not.

Let's make 2026 remarkable.

**Happy New Year, builders. Let's do this\! **

## Read more
- [5 Founder-Approved Tips for a Christmas Reset](https://videotranslator.ai/blog/5-founder-approved-tips-for-a-christmas-reset/)
- [It was Supposed to Happen in 2027. It Arrived Early.](https://videotranslator.ai/blog/it-was-supposed-to-happen-in-2027-it-arrived-early./)
- [2025's AI Kaleidoscope. Did You Miss This?](https://videotranslator.ai/blog/2025s-ai-kaleidoscope-did-you-miss-this/)


Read I'm Done Pivoting. Here's What I'm Building in 2026. and explore practical insights from VideoTranslatorAI on multilingual communication, interpretation, and AI-supported workflows.

I'm Done Pivoting. Here's What I'm Building in 2026.

You have permission to close your laptop this holiday season.

I know, I know. That sentence probably triggered your founder anxiety. 

*But what about the roadmap? The bug reports? The competitor who just shipped?*

Before your anxiety kicks further, let me tell you something:

**“The founders who burn out don't lose to competitors who work harder. They lose to competitors who recover smarter.”**

I learned this the painful way. Two years ago, I convinced myself that “staying ahead” meant working straight through the entire holiday break. 

I was heads-down shipping new features, polishing my app, and frantically putting together all the docs for the second phase of the [NSW Small Business Innovation and Research (SBIR)](https://www.chiefscientist.nsw.gov.au/news/nsw-sbir-program-awards-$1-million-in-grants-to-small-business-to-tackle-nsw-problems) program.

No real days off, just me telling myself it would give me a massive head start in the new year.

By January 15th, I was so fried that I made a product decision that took months to undo.

Then, a random post popped up on my X timeline: **“Strategic rest is a force multiplier.”**

Those words hit me hard. I realised I couldn’t keep running on empty if I wanted to build something sustainable.

So, during last year’s holiday season, I did something radical for me: I took two full weeks off. I actually closed the office, gave my small team proper time to recharge, and unplugged as much as possible.

The difference was night and day. I came back clearer, sharper, and honestly more creative.

As my Christmas gift to you, I’m sharing all the **simple practices** I now use to protect my energy over the holidays, without letting my app or business suffer when January hits.

## 1\. Set Boundaries That Actually Work

![5 Founder-Approved Tips for a Christmas Reset - Blog](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/b2f95146a2896d100cd4090f7c2c336ddc08ba7e8c3824667d187da018461577.webp)

You wouldn't ship a feature without defining its scope. Apply that same rigour to your rest.

**Block specific hours** in your calendar as non-negotiable downtime. I use 7 PM to 10 AM during the holiday week. These aren't suggestions. They're requirements.

**Communicate clearly** with your team and users. A simple auto-reply works: *"Recharging through the 26th. Back online soon."* People respect founders who model healthy boundaries.

**And remember**, the benefit of proper breaks isn't just "feeling rested." It's cognitive. Your brain solves problems differently when you're not actively forcing it. 

I've had more product breakthroughs during walks or conversations with non-tech friends than during focused work sessions.

Rest isn't the absence of productivity. It's a different kind of productivity.

## 2\. Micro Self-Care Rituals (That Don't Feel Indulgent)

![Photo by Yosuke Ota on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/d38992800262bf54befd2d97e593626a9d638ef8d85e04fbc490c8bc8e1bbbc6.webp)

Self-care sounds like spa days and expensive retreats. For most founders, it's simpler: small rituals that reset your nervous system.

Here's my holiday stack:

* **Screen-free mornings**: First 45 minutes of each day. No devices. Just coffee, read physical books or talk to my partner.   
* **One real conversation**: Fifteen minutes of genuine, phone-down connection with someone I care about.  
* **The guilt-free movie**: That film you've wanted to watch for six months? This is the week.  
* **Five minutes outside**: Fresh air resets your nervous system faster than any app.

None of these requires planning or money. They simply require you to take a few minutes for yourself.

## 3\. The "Done List" Instead of To-Do List
![Photo by Andrew Neel on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/47e5ff476d5984e4a161c1bc9633cb8a7896dfcb7c55ecddbf69cad1b687e58a.webp)

Founders obsess over what's not finished. The features you didn't ship. The revenue targets you didn't hit. The "never enough" mindset that makes every achievement feel temporary.

Counter this with a "**Done List.**" Sit down, open a document, and write every single thing you achieved in 2025\. 

Not just launches or revenue milestones. Everything.

Shipped that bug fix nobody noticed but you? Write it down. 

Had that difficult conversation with a team member? Write it down. 

Learned a new framework, rebuilt your landing page twice, and survived a server outage at 2 am? Write it all down.

By the time you finish, you'll have pages. You'll remember that you actually built a lot this year, even if it doesn't feel like it. 

Keep this list somewhere visible. When January's imposter syndrome kicks in, read it.

## 4\. Pre-Clear Your January Runway
![Photo by Debby Hudson on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/3edab1798ae3657342b646bed19a7638bc573c0930fa1f84daac0a30e9b7e3de.webp)

January you will thank December you for this.

Spend 30 minutes before the break listing every low-energy task you can batch: updating documentation, responding to non-urgent emails, scheduling social posts, organising your task manager, and backing up important files.

Do some now. Schedule the rest for specific January days. 

The goal isn't finishing everything. It's removing the mental load of "remembering to do this later."

I use the last working day of December to set up my January calendar structure. Block times for deep work. Schedule the boring admin tasks. Pre-write messages to team members about priorities.

Starting January refreshed is useless if you start it disorganised. Ten minutes of setup now saves hours of "where do I even start" paralysis in January.

## 5\. Schedule One "Weird Reset" Activity
![Photo by Jasmin Schreiber on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/685c41f32d2d9a280eef05df1810f4a9cb0208d6352f89c223ebbdc77d529d49.webp)

Here's my unconventional tip: **do something completely unrelated to tech that uses your hands.**

Bake bread. Build a LEGO set. Fix something broken around the house. Go for a hike. Take a pottery class.

Why this works? Because founders spend all day in abstract problem-solving.

Tangible, physical activities activate different neural pathways and give your "strategy brain" a genuine break. 

I built a bookshelf last December. Terrible craftsmanship. Best reset I've ever had.

## What I'm Reminding Myself This Year

Building a company is a marathon, not a sprint. 

Everyone says this, but nobody explains what it means practically.

**It means: burning out in December 2025 makes you less effective in March 2026\.** Skipping rest now doesn't accelerate success. It delays it, because you'll hit a wall later when it's harder to recover.

**It means: your app will still be there on January 1st.** The features can wait. The users can wait. The market can wait. You, however, cannot run indefinitely on depleted energy and expect to make good decisions.

**It means: taking care of yourself isn't selfish founder behaviour.** It's responsible founder behaviour. You're the most important infrastructure your startup has. If you break, everything breaks.

## Merry Christmas, Builders
![Photo by Tessa Rampersad on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/0aa6274c040ea8254b14a6c05468eabee986e29f2b7618d947a79757275ea0e5.webp)

To everyone building something this year: you did more than you realise. You survived moments that felt impossible. You kept going when giving up seemed easier.

Take the break. Actually take it. The work will be there when you return, and you'll be better equipped to handle it.

Merry Christmas. Happy holidays. Whatever you celebrate or don't celebrate, I hope you get some genuine rest, some good food, and some reminder that you're more than your startup metrics.

You're building something meaningful. That takes time. That takes energy. That takes breaks.

See you in 2026\. We've got more to build\!

## Read more
- [It was Supposed to Happen in 2027. It Arrived Early.](https://videotranslator.ai/blog/it-was-supposed-to-happen-in-2027-it-arrived-early./)
- [2025's AI Kaleidoscope. Did You Miss This?](https://videotranslator.ai/blog/2025s-ai-kaleidoscope-did-you-miss-this/)
- [The Reading Advantage No AI Can Replicate. Build It Before 2026.](https://videotranslator.ai/blog/the-reading-advantage-no-ai-can-replicate/)

Clients aren’t committing. Your runway is shrinking. Relax over the holidays anyway?

5 Founder-Approved Tips for a Christmas Reset

In 2020, asking an AI to translate **"*I'm feeling blue*"** would likely get you something about the colour of your skin. 

Or when you translate ***"that joke killed"***, it might suggest something about death rather than success.

Five years ago, AI translation was technically impressive but practically awkward. 

But today, you point your phone at the menu and read it in perfect English. Or you ask the chef a question in English and hear their response in English, while they hear your question in Japanese.

The transformation happened so fast that most people missed the individual breakthroughs that got us here.

So, before 2025 ends, let me walk you through what actually changed, year by year, and why it matters more than you think.

## 2020: The Year Machines Learned to Think in Sentences
![Photo by Andrew Neel on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/f0c4ffde3c912c7881fc211534d7e1320ae46231b11c63bb1ae06f2962d25c49.webp)

In May 2020, OpenAI released [GPT-3](https://developer.nvidia.com/blog/openai-presents-gpt-3-a-175-billion-parameters-language-model/), a language model with 175 billion parameters. That number itself meant nothing to most people, but what it could do was remarkable: [generate text that sounded genuinely human.](https://arxiv.org/abs/2005.14165)

GPT-3 represented a breakthrough in natural language processing because it dramatically increased the quality and accuracy of language models. For translation specifically, it meant AI could finally handle longer sentences and complex ideas without falling apart halfway through.

Before GPT-3, AI translation worked on smaller chunks of text. Feed it a paragraph, and it might lose the thread halfway through. 

GPT-3 could maintain context across much longer passages, understanding how the beginning of a sentence related to the end, how paragraphs connected to each other, and how tone should remain consistent throughout.

The practical difference? Translations started sounding less like a robot trying to mimic human speech and more like an actual person translating thoughtfully. 

Not perfect, but recognisably human-like for the first time.

## 2021: Smaller Languages Finally Got Attention
![Photo by Tien Vu Ngoc on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/97e1425c2e1e84cdbee2e04ae3e38092687dc3f0ef6d4100397dc818df35b3cd.webp)

Machine translation had always played favourites. English, Spanish, Mandarin, and French: these languages got the attention and the quality. Everyone else got scraps.

2021 marked a shift toward multilingual models that shared knowledge across languages. 

Instead of training separate models for each language pair (English-French, English-Spanish, etc.), researchers developed models that understood many languages simultaneously.

What the model learned about grammar in one language helped it understand patterns in others.

For the first time, speakers of smaller languages started seeing translations that actually worked. 

Not perfect, but functional. Not "*language not supported,*" but "*here's what they're saying.*"

## 2022: No Language Left Behind (Literally)
![Source- Meta](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/7f8a275d341c22dd61d1c440b50d836e147c481145fdcf86f6d6183a4a65aae6.webp)

In July 2022, Meta AI announced an ambitious project: [the No Language Left Behind (NLLB-200)](https://ai.meta.com/research/no-language-left-behind/), a single AI model capable of translating across 200 different languages with state-of-the-art quality.

This wasn't just an incremental improvement. This was a fundamental shift in what AI translation could cover. 

Before NLLB-200, fewer than 25 African languages were supported by widely used translation tools, and most of those poorly. NLLB-200 supported 55 African languages with high-quality results.

Microsoft took a complementary approach with its [Z-code Mixture of Experts models](https://news.microsoft.com/source/features/ai/new-z-code-mixture-of-experts-models-improve-quality-efficiency-in-translator-and-azure-ai/). This model focused on improving both quality and efficiency in translation. 

Rather than using one massive model for everything, [Z-code](https://techcrunch.com/2022/03/22/microsoft-improves-its-ai-translations-with-z-code/#:~:text=Microsoft%20today%20announced%20an%20update,Image%20Credits:%20Microsoft) trained specialised sub-models that activated only when needed for specific language pairs or contexts. This made translation faster and more accurate whilst using fewer computational resources.

![Read about Z-code Mixture of Expert here](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/555bdbeb8c2a483c50ff7db85e3f46df31a16259400d7961791c25fe6331a7aa.webp)

The combination of Meta's NLLB-200 pushing breadth (supporting 200 languages) and Microsoft's Z-code improving depth (better quality per language pair) represented a two-pronged attack on translation limitations.

This was the year translation became genuinely democratic. 

It was *recognition*. It was *access*. It was finally being able to participate in the global digital conversation.

## 2023: Context Becomes Everything

![Photo by Aerps.com on Unsplash](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/9283a6cbdf2f85fa3d24ad3229b5d7c2c866203465af2c68a4f8e6b8372981be.webp)

Then came the integration with conversational AI. Models like ChatGPT and GPT-4 brought context-aware, instruction-following capabilities to translation.

Before 2023, translation was binary. You put text in, you get text out. You had no control over how it sounded.

The integration of translation with conversational AI changed everything. Suddenly, you could tell the AI exactly what you wanted – tone, style, formality level, or intended audience. 

 ***"Translate this into French for a new client. Keep it friendly but professional."***

This solved a problem human translators have always understood, but machine translation struggled with: translation isn't just swapping words between languages. It's carrying meaning, intent, and tone across cultural and linguistic boundaries.

A phrase like "*I hope this email finds you well*" translates literally into many languages, but sounds odd or overly formal in some contexts. 

Context-aware AI could now adjust that opening to match local business communication norms rather than just doing word-for-word translation.

For users, this meant steering translations: formal versus casual, simple versus detailed, technical versus accessible. Translation became a dialogue with the AI, not just input-output.

<iframe width="1236" height="695" src="https://www.youtube.com/embed/dHxOf3Tf9vo" title="Designing for Vibe, Not Just Translation" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

## 2024: Finally Getting The Joke

![Photo by Vitaly Gariev on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/164d0e5da51a8c039030fffe93c7bd6c800315821a2ab871aa0f0c792f136db4.webp)

"*That presentation absolutely killed*."

To a human, this is obviously positive. To earlier AI systems, someone may have died.

2024 brought major advances in understanding idioms, sarcasm, and emotional context. 

AI learned that "*break a leg*" is encouragement, not violence. That "*it's raining cats and dogs*" describes weather, not falling animals. That tone matters as much as words.

Translations started preserving not just meaning but feeling. The warmth in a message stayed warm. The urgency stayed urgent. The humour, finally, stayed funny.

![ChatGPT joke translation.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/1587739d942bde721c7d33100e5b9214800a79fc194399f11d0e386ca68b658e.webp)

## 2025: The Star Trek Dream Becomes Real

![Source- Apple](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/39bdf207cf08a295798679816ef4109dd29cf6d64022e4c71534fa0d97046167.webp)

Remember the universal translator from the famous Star Trek movie? That one device that lets characters speak any language instantly?

In 2025, it arrived. In your pocket. In your ears. Even on your glasses.

[Apple's AirPods](https://www.bbc.com/travel/article/20251001-how-real-time-translation-could-transform-travel) now offer real-time translation. Speak in your language, and the person across from you hears their language. They respond, and you hear yours. The conversation flows naturally, as if the language barrier simply didn't exist.

[Google Meet](https://blog.google/products/workspace/google-meet-langauge-translation-ai/) brought the same magic to video calls, with AI that preserves not just words but the speaker's tone and emotional expression. Your voice, your meaning, their language.

Even smart glasses from [Ray-Ban](https://www.meta.com/ai-glasses/meta-ray-ban-display/) can now translate the world as you look at it. Signs, menus, conversations: all accessible, all instant.

Some of these tools work offline. The translator lives on your device, ready whenever you need it.

## What This Means for You (and Everyone)

Let's be concrete about what these five years of progress actually enable:

A small business owner in Vietnam can now have video calls with potential clients in Brazil, each speaking their native language, and actually understand each other naturally. A few years ago, this would have required hiring interpreters or struggling through awkward, expensive translation services.

A traveller in rural Japan can have conversations with locals using their phone, getting recommendations, asking directions, understanding cultural context – all without speaking Japanese. Previously, you'd be limited to tourist areas where English is common or rely on phrase books and gestures.

A student in Kenya can access educational materials written in English, French, or Mandarin, translated into Swahili with cultural context preserved and idioms properly adapted. Before NLLB-200, many educational resources simply weren't available in smaller languages.

A healthcare worker can communicate with patients who speak different languages during emergencies, getting critical information about symptoms, medications, and allergies accurately and immediately. Language barriers in medical settings can literally cost lives. AI translation is starting to address this.

These aren't future scenarios. This is happening now, enabled by the cumulative advances from 2020 through 2025\.

## How This Shaped Our Company

These developments didn't just benefit users. They fundamentally changed what we could build.

When we started developing AI-powered video translation services, we were working with technology that could generate captions from video content but struggled with translating those captions *naturally* across multiple languages. 

The AI could transcribe what was said, but translating it whilst preserving tone, handling idioms, and maintaining timing synchronisation with the video was incredibly challenging.

![Our early video translator days](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/dc6fe84630040709a601f320467d54bbf947101de8c00393e0944a4d9f632f0f.webp)

As each breakthrough happened from 2020 onward, we rebuilt and expanded our platform to leverage these advances. 

GPT-3's release meant we could offer more natural-sounding translations. The multilingual models of 2021 let us support a broader range of languages. NLLB-200 in 2022 enabled us to serve communities speaking minority languages who'd been neglected by mainstream tools.

The context-aware capabilities allowed us to build features where users could specify the tone and formality they needed. When idiom and tone handling improved in 2024, our translation quality for casual speech and culturally-specific content jumped dramatically.

And in 2025, we're now developing real-time speech translation for meetings and calls, with transcription, translation, and downloadable summaries in multiple languages. None of this would have been possible with 2020 technology.

![Italian-English Agent Interpreter.gif](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/722aaaea7b6fd89c483167339c915daba98848cb5ed5bfa4386d81201d88718f.gif)

The five-year progression didn't just make existing features better. It made entirely new capabilities viable. 

Features we couldn't have built in 2020 are now standard in our platform, and we're developing things that would have seemed impossible just two years ago.

## What Comes Next

![Photo by Tom Parkes on Unsplash.webp](https://storage.googleapis.com/vtai-web-media-prod/cms/originals/blog-inline/77d3c33264abd93cb7be46bc7e3cd7102c7d5cb187f7d371e088c24b6f400a5b.webp)

We're at an interesting inflection point. AI translation has gone from "technically impressive but practically limited" to "genuinely useful for real-world communication" in just five years.

The remaining challenges aren't about basic capability anymore. They're about refinement: reducing the last instances of cultural misunderstanding, improving emotional tone preservation, handling highly technical or specialised content, and making real-time translation work smoothly even in challenging audio conditions.

Each year's breakthrough built on the previous year's foundation. GPT-3's language understanding enabled better multilingual models. Multilingual models made NLLB-200 possible. NLLB-200's scale enabled context-aware translation. Context awareness enabled proper idiom handling. All of this together made real-time voice translation viable.

The trajectory is clear: language barriers are becoming increasingly optional rather than absolute. 

You still might prefer a human interpreter for crucial negotiations or sensitive conversations. But for everyday communication, travel, business calls, accessing information, and connecting with people globally, AI translation is reaching the point where it just works.

Five years changed everything. The next five will probably change everything again.

But right now, take a moment to appreciate how remarkable it is that you can have a conversation with someone speaking a completely different language, in real-time, using a device in your pocket.

That's not science fiction. That's 2025\. And it's pretty incredible.

## Read more
- [2025's AI Kaleidoscope. Did You Miss This?](https://videotranslator.ai/blog/2025s-ai-kaleidoscope-did-you-miss-this/)
- [The Reading Advantage No AI Can Replicate. Build It Before 2026.](https://videotranslator.ai/blog/the-reading-advantage-no-ai-can-replicate/)
- [In OZ? 23 Courses You Can Now Take (Legally) to Get Hired Faster](https://videotranslator.ai/blog/in-oz-23-courses-you-can-now-take-legally-to-get-hired-faster/)

A year-end look at the technology that brings us closer.

It was Supposed to Happen in 2027. It Arrived Early.

We use cookies to improve site performance and understand how the portfolio content is being used.

Elephant Stripes is a consulting firm that designs, builds, and ships AI products and delivery systems.