In July 2025, Microsoft researchers published a study that aimed to quantify the “AI applicability” of various occupations. In other words, it was an attempt to calculate which jobs generative AI could do best. At the very top of the list: Translators and interpreters. The paper itself was strange (historians and passenger attendants took the second and third place slots) but it underlined a talking point that’s been roundly discussed in the media: That translation work is uniquely vulnerable to AI.

To wit: After I put out the call for AI Killed My Job stories, I heard from a lot of translators, interpreters, and video game localizers (essentially translators for in-game text, design and dialogue). Of all the groups I heard from, translators had some of the most harrowing, and saddest, stories to share. Their accounts were quite different from those described by tech workers, who were more likely to lament managements’ overuse of AI, a surfeit of dubious code in digital infrastructure, hasty layoffs, or the prospect of early retirement.

AI Killed My Job: Tech workers Brian Merchant · Jun 25 Read full story

For most translators, early retirement is unthinkable. Many of the translators I heard from were underpaid and precariously employed before the AI boom hit, and had stuck with the field because they loved the work despite the downsides. Now, as you’ll see in the stories below, many have seen truly dramatic drops in their income. Multiple accounts describe work drying up almost entirely, and the prospect of having to change careers at a time when peers in their age group are thinking about retirement.

I also heard from a lot of game localizers working with Chinese mobile games in particular, perhaps because it’s an industry that touches both media and tech, where leadership may be more disposed to embrace AI initiatives. I received too many to include them all here, but suffice to say, the stories almost all described games companies drastically lowering rates, increasing reliance on AI for translation (with or without human editors), and slashing in-house localization staff.

In an interesting—and rather telling—wrinkle to the AI boom story, many translators noted that generative AI didn’t usher in any revolutionary improvement to already-existing technologies that have been used to automate translation for years. Long before AI became the toast of Silicon Valley, corporate clients had been pushing lower-paying machine translation post-editing (MTPE) jobs¹, or editing the output of AI translation systems, though many translators refused to take them. Others said Google Translate had long been able to essentially what ChatGPT does now.

Yet many describe a dramatic disruption in wages and working conditions over the last two years, coinciding with the rise of OpenAI. Though my sample size is small, these stories fit my thesis that the real AI jobs crisis is that the drumbeat, marketing, and pop culture of "powerful AI” encourages and permits management to replace or degrade jobs they might not otherwise have. More important than the technological change, perhaps, is the change in a social permission structure.

Not one but two accounts detail how many translators dismissed ChatGPT at first, because they’ve heard companies tout many automation technologies over the years, all with limited impact—only to see the floor drop out now. And it’s not that ChatGPT is light years better than previous systems (lots of post-AI translation editing is still required), it’s just that businesses have been hearing months of hype and pontification about the arrival of AGI and mass automation, which has created the cover necessary to justify slashing rates and accepting “good enough” automation output for video games and media products. Everyone else is doing it, after all.

Yet much stands to be lost, even aside from decent wages and the livelihoods of the translators and interpreters who help make our cultures better understood. The quality of translations across the board, from video games to corporate communiques stands to decline, with AI output, according to interviewees, often being homogeneous, blind to local details, or flat-out wrong. Nuances about places and cultures, recognizable to a knowledgeable human interpreter risk disappearing, sanded down by blunt-force automation. It’s not overly dramatic to say that we risk losing the capacity for cultures to understand one another better if we’re all simply feeding output into each other’s automated translation systems.

These risks are existential enough that groups are organizing to push back. The Translators Against the Machine initiative is gathering stories and data about what it’s like to work in the industry right now, in a bid to grow solidarity among far-flung workers, and to “unite and join forces to rescue the translation profession from the claws of a market that aims to make us irrelevant and expendable.” The English-to-French games translator Lucile Danilov, who we’ll hear from shortly, has worked to poke holes in the ways that AI companies have been pitching AI translation to games companies. Forums and message boards are seething with discontent.

It’s of course unclear what the future holds, but there’s a growing sense that the AI phenomenon is more bubble than boom. As such, rather than viewing the enterprise AI frenzy on Silicon Valley’s terms, as an inevitable jobs apocalypse, we have an opportunity to view it on material terms, and examine how it’s actually playing out on the ground. On those terms, we see managers, executives, and corporations using rebranded automation software to increase volume and cut labor costs, starting with the most precarious workers. After all, an AI system does not have to be super-powerful for management to use it to degrade, deskill, and kill jobs. This, it seems, is what translators, interpreters, and localizers are experiencing, right now, on the front lines of the real AI jobs crisis. And these are their stories.

Three very quick notes before we move on. First, this newsletter, and projects like AI Killed My Job, require a lot of work to produce. If you find this valuable, please consider becoming a paid subscriber. With enough support, I can expand such projects with human editors, researchers, and even artists—like Koren Shadmi, who I was able to pay a small fee for the 100% human-generated art above, and Mike Pearl, who is coming on to help edit installments in this project. If you would like an alternate way to offer support, I now have a Ko-fi page. Many thanks.

Second, if your job has been impacted by AI, and you would like to share your story as part of this project, please do so at AIkilledmyjob@pm.me. I would love to hear your account—and will keep it confidential as I would any source. Third: I'm partnering with the good folks at More Perfect Union to produce a video edition of AI Killed My Job. If you're interested in participating, or are willing to sit for an on camera interview to discuss how AI has impacted your livelihood, please reach out. Thanks for reading, human, and an extra thanks to all those whose support makes this work possible. I have countless more stories in fields from law to journalism to customer service to art to share. Stay tuned, and onwards.

Translators have always been among the most threatened by automation

Translators have always been among the professions most threatened by automation, long before the advent of AI, through the development of machine translation engines like Google Translate or DeepL. But in recent years, the situation has dramatically worsened, despite LLMs producing consistently mediocre results.

The very definition of translation is not to convert words, but meaning. And while LLMs are able to replicate human speech patterns with eerie accuracy, it bears reminding that they don’t think nor understand like the human brain does. Which means that editing LLM outputs often takes as much time, if not longer, than translating from scratch.

Despite that, in an effort to cut costs and lower turnaround times, many translation agencies have been increasingly switching to a business model revolving around MTPE (Machine Translation Post Editing), slashing rates and often compromising the quality of the final product. This practice has long been seen as a bane by most translation professionals, who feel like their skills amount to a lot more than mere word-assembly lines to target the lowest common denominator.

Now, the concept of “polishing” a machine output is bleeding across all industries, and many are starting to realize that translators were the proverbial canaries in the creative coal mines.

-Lucile Danilov

Terrible Google translations once made the idea of automated translators laughable. I’m not laughing anymore

I have been using Computer-Aided Translation (CAT) tools for the past twenty-five years, as translation has always been an area of focus for machine learning and programming. The people creating these programs have heralded the end of human translation since the 1950's, with the Georgetown-IBM experiment in 1954. Back then they thought that it would just take a few years for machines to take over. What a joke!

Since that did not happen for decades, most of us in the translation industry scoffed at the idea that computational linguistics would ever find a way to replace us. I indeed use machine translation in my practice, much like an accountant uses software or an airline pilot uses autopilot. These tools are meant to take over routine tasks and reduce fatigue. (Although, even when using these tools, I still suffer from back and neck issues from sitting at a computer all day.) They help, but the idea that they can replace a human translator is ridiculous.

With ChatGPT, I honestly didn't think it would change anything and that people would still think of it with disdain as they did Google Translate (at least, they do where I live). But for the past year or so, I noticed that many clients have had less volume. When I asked them about whether they had work for me, people would say that of course I was first on their list, but they just didn't have anything. For the past six months, I have seen translators on LinkedIn that I know say that the same thing is happening to them. The work has dried up.

While I didn't have any real proof that my clients were choosing ChatGPT over me, a recent project showed that they are using this (or DeepL or some other AI tool). I was asked to translate a large document for an existing client I hadn't heard from in a while. They told me to refer to their website for their terminology. I said okay, wondering who exactly had been translating their site. When I read their site, you could see all of the telltale signs of AI: acronyms and terms translated inconsistently (this is a big sign), as well as the usual weird constructions and some of it just plain nonsense.

So now this creates work for me: I have to somehow refer to and use this slop while still doing a professional job. Then, when I have no choice but to change it, I have to write nice, diplomatic notes about the change!

I also learned in this big document about all of the content they produce to communicate with their audience: bulletins, emails, web content. And I think, They cannot be using AI for all of this? The result must be bad. But, according to their statistics, they still have a 50% open rate on their emails and people don't unsubscribe. (This is about the same as when I was translating their emails.) So, are they getting a cheaper translator or in-house staff to work on this content? Does the audience just not care that the translations are bad? I think, to some extent, people are inured to bad translations. "Well, I guess it's in English, so that's better than nothing, whatever." I don't know.

Another thing that seems to be happening (based on my anecdotal experience only) is that translation agencies are investing in this tech and then gobbling up the work from freelancers. And translation agencies really don't pay well, and even less for what we call "post-editing" (a fancy term for "fix the machine"). I recently put in a quote to a regular client for a contract at about 2/3 of the regular price that I charge, and they said that I was still too expensive. As a comparison, during the pandemic, I had to raise my prices as I was so busy. And so, now what, I have to lower them? What message does that send to clients about the value of my work?

Add the tariffs to this mix, and my economic life has completely turned upside down. No one wants to invest in new initiatives and projects. I do have diverse revenue streams, as I teach English online and edit novels, but teaching and editing do not pay nearly what translation does. I will be lucky to make $15 to $20 an hour teaching and editing, when translation pays at least 2 to 3 times that. So here I am, still 15 to 20 years from retirement, and I have had to put so much energy into "reskilling" for teaching and editing, which, even if they don't pay as well, are jobs that require in-depth, professional levels of skill! I laugh when politicians think that "reskilling" is this magical thing that anyone can pick up and do. Going back to school in your 40s and 50s when you still have kids to take care of, a house to manage, finances to manage, health to manage, aging parents to worry about and take care of, etc., etc. is one of the biggest jokes on working people. Fuck, am I tired.

I do count myself lucky, as even though my income has been cut in about half, I still have enough to live on for the moment. I have health insurance and retirement benefits through my province. (I also got a year of maternity leave, even though I am a freelancer, thank goodness.) I have no idea how people in the US without this safety net can get by without these essential programs.

However, it's not like simply living on less is a great option either. Housing in Canada is extraordinarily expensive. Food is expensive. And my employment is very precarious. I always have to plan for the day when the worst will happen. I can never relax. Living in this AI-driven economy is always expecting the rug to be pulled out from underneath you. I invested in my profession for years, and now I don't get to reap the benefits of that investment. I have to start all over again, always putting energy into this bottomless pit they call work.

For me, AI means white knuckling it your entire life until you retire. What joy, what rapture unforeseen!

-Anonymous

AI killed my job twice, maybe three times

AI killed my job. I think I can even say it's killed my job twice (possibly three times!?). With more to come!??

I graduated from my translation MA in 2010. I was in-house for a few years and then went freelance and was doing quite well—I was always an early adopter of new tech so was one of the first to take MTPE work. I saw that change my dynamic and then have seen it happen again once I sidestepped into copywriting around 2023, just as AI was really ramping up. I helped train a model for a big company... And then they got rid of me!

I've now done yet another pivot and I'm working in email building, using tools like Klaviyo and Braze, but I imagine that's vulnerable too.

I'm only 40. I never imagined my career as I knew it would be wiped out like this.

-Anonymous

I’ve translated documents for nuclear power plants. Now I’m facing bankruptcy

I've been a technical translator for 15 years, self-employed all the way. I enjoy it, I am good at it. I translate complicated, demanding material—mainly medical and pharmaceutical, like the UI and user guides for MRT imaging devices, or patient information and consent forms for clinical trials, or subtitles for a presentation on the side-effects of this or that new drug. I've translated documentation for the specialty filters you need in cooling loops for nuclear power plants and I've translated manuals for assembly systems for aircraft construction. I get to dive into obscure sub-specialties of technical fields and learn about stunning feats of engineering nobody has ever heard of. It's fun. In a field where everybody seemed perpetually on the brink of starvation, I was able to make a good living. There were always ups and downs, but I managed to clear six figures in the good years and didn't have to worry too much in the bad years. I worked long hours, I worked a lot of weekends, but I felt it all balanced out.

2025 has been absolute shit so far. Entire months went by with zero work. And the requests that are now coming in—almost all "PED.” Post-editing is when you run your text through a machine translation and have it reviewed and edited by a human. It's been around forever—since way before the current AI hype. It pays a quarter of what you'd get for translation work. And if you do it properly, it takes you just as long as translation. So I would summarily reject PED requests. I'd take one or two per year just to take a look at the current state of the art, and invariably found, happily, that machine translation was still awful and I was going to be fine.

As of today, I've earned maybe €8000 [about $9,300] this year. Requests are 90% PED. Unrelated calamities have drained the vast majority of my savings (just lucky, I guess). There is a very real possibility I'll end up in personal bankruptcy.

Machine translation hasn't even improved. There was no big OpenAI moment. I'm starting to suspect it's an unhappy coincidence of sunk costs and economic downturn forcing us all down this path. And you know what? I started learning to code—needed something to do after all. And ChatGPT and Claude started off as amazing helpful tools. Then at some point you've got the basics down and you're trying to do marginally more complex things—and you notice how quickly they lose track and fall apart, how needlessly complicated their solutions are, how your entire architecture turns into a mess of barely-functional spaghetti. Does this stuff work *anywhere*? My IT friends complain about being forced to use whatever hot new AI tool, and their companies stopped hiring junior positions. My own industry seems broken. After sending this mail, I'll have to do some tedious, underpaid post-editing. I'll hate it. Whoever will have to actually use the documents will hate it.

I believe this will pass. Sooner or later, the AI companies will have to stop losing money and adjust their pricing. And then it'll turn out that using AI for everything gets you worse results than humans, at the same cost. And that will be that. I hope I can hang on until then.

-Julian Pintat

A brain drain is coming

I'm a translator trainer at the University of Geneva, training people to work at the UN, WHO etc. One of our big challenges is getting young people through the doors to train—there is likely a big talent gap coming at the top of the profession as people retire over the next decade or so.

-Susan Pickford

I was a different kind of translator, but AI hollowed out the work

I was working as an accessible information writer. We would translate technical documents into Plain language (think gov sites) or instructions into Easy English (think “How to Catch a Train” for people with intellectual disabilities).

Although AI is expressly banned from being used to actually write the documents, AI was being used to check the documents, and then those modifications had to be used to re-edit those documents.

I’m not sure if management realized they were getting AI to write these documents—but with extra steps—or if they thought they were somehow bypassing internal policy, or if they thought this maintained privacy. It was quite plain to me that this workflow is not doing any of those things.

I left the job recently because I could see where it was going. Also because this was a top down initiative it was causing friction in the team. Writers were essentially being told to write for AI, then let the AI take the reigns.

This might seem like, sure, why not turn up to work and take the free money? But it was actually causing massive issues. Writers were being put on notice when our documents were being checked by peer review. No one on the peer review team would agree to a final copy. And so the sausage was fed back into the machine only to be stopped at peer review again. Then the writer was held accountable.

Even though AI was not directly being used to write the documents because it was in the middle of the process it may as well have been used. The outcome was unusable work for which the writers were being blamed. Sad stuff.

-”FF”

AI-only translation is happening now

I had been working on a mobile game for years and the agency that manages them recently told me the game will only have AI translation with no human proofreading, for my language pair and many others.

Generative AI in games is no surprise, most devs that have no budget resort to DeepL or Google, but the shift that I’ve seen is coming from big players and games that do earn money and are able to play localization: they just don’t want to anymore. Plus, they lower your rates so when the MTPE comes, you view it as you have no choice, because it’s either that or not working at all.

We’re at the expense of rich folks that want to be in the loop, discrediting our job while they do not know what it entails.

-Tamara Morales

After 14 years of translating to English in Rome, I’m considering cleaning houses

I'm an Italian to English translator living in Rome. I've done this job freelance for 14 years now, before that I worked in the music industry at a startup in SF. In these past 14 years as a translator, I've worked hard but I've also learned a lot while doing something I deeply love. Some of my clients, mostly agencies, started asking about MTPE a few years ago, and I told them that revising a machine translation takes me longer than a translation from scratch so why should I accept half my rate for it? Some of them started offering MTPE to their clients, but not all of them, and many of the translations I work on aren't really suitable for machine translation. I didn't really see a drop in work at that point, actually I had my best year ever in 2024. I diversified into copywriting as well.

Fast forward to June 2025. I did not receive a single work request AT ALL that month. I went from working 50-60 hours a week to essentially working zero. This month (August 2025) some work has trickled in, but it's very sporadic and unreliable, meaning I can't really do this job anymore and expect to pay my bills.

I don't know what to do at this point. I'm 44 years old, I've already changed careers in my lifetime and the job market is terrible, despite my experience as a translator, operations manager, degree in art history, fluent in 2 languages, decent and one and learning yet another. It feels like I might as well just start cleaning houses for a living, at least that's steady work and hasn't been replaced by AI yet. I have to wonder: once they've pushed us all out of our jobs, who will have the money to buy the products and services that capitalism requires of us?

-Katherine Kirby

We’re being paid half as much to do lower-skilled work

I have formal training in translation and have been working in the translation industry for 15 years, 5 years as a translator and 10 as a translation project manager. I'm 40 at the moment.

Work has been depressing, to say the least. All the projects I receive are AI translated and many of the translators I work with complain about the quality and the lack of work

The clients don't care, all they see is a cheap way to translate stuff and the faster, the better. Translators are now post editors or reviewers. Quality in translations has been decreasing but no one seems to care.

I work part-time as a freelance project manager and have been trying to get some freelance translation jobs on the side. All the job posts I see are for "AI trainer", "AI translation assistant", "AI assisted translator", etc.

It's disheartening. My industry has always been underpaid, for my language pair (English-European Portuguese), the medium rate is 0.04 € per source word. Now it's 0.02 € per source word for post editing AI translations. Many translators are accepting these rates because otherwise they would earn nothing.

I'm mentally exhausted just thinking about this. I want to change jobs, I want to work with something that will not involve computers and AI because I fear many jobs will be killed by AI.

-Anonymous

My work in corporate communications has come to a complete stop

I’ve been a freelance French-English translator since 1997, working primarily in corporate communications for large companies in France. My work started gradually diminishing about two years ago but has come to a complete stop this year and I’m having to find other sources of income.

I have always worked primarily for translation agencies that serve large companies and subcontract the actual translation work to me, although I have (or had) some direct clients as well. Machine translation has been around within the translation industry for many years now on basically the same basis as today’s AI.

Translation customers were aware of that, but since the translation service providers maintained the memories, they retained the upper hand, while freelance translators like me were mostly relegated to editing the computer output (a tedious task that never paid as well as translating). Now I suppose companies have realized they don’t need the outside provider and can just feed their text into a program like DeepL. About 5-7 years ago I started spending a far greater share of my time on editing computer output rather than translating, but my overall volume of work stayed roughly the same until two years ago. Now I’m getting virtually nothing. It’s certainly very rare now that I get a request to simply translate a document.

I’m 62. Translation has never been a high-paying career (my rates have barely changed since 1997; there’s intense downward pressure on rates, partly because competition is global) and I planned to continue working until I was nearly 70, but this has been a very disruptive change—at my age it’s very difficult to start a new career or even get hired.

-Anonymous

In 2019, companies would reach out to me. Today, I’m the one reaching out—and often being ignored.

I’m a 32-year-old from Italy, [and] I think that in the U.S., people are underestimating the impact AI is having on the millions of remote workers worldwide who, for over 15 years, have been silently doing much of the behind-the-scenes work for tech companies. I personally know hundreds of remote workers from Europe, Asia, and South America who are now struggling because of AI: Spanish translators from South America, low-level programmers from India, editing and graphic design experts from South Asia… Why hire them when AI can now do 95% of their job?

But let's go back to my personal situation: I studied History at university, but as you can imagine, finding a job related to that field in Italy proved nearly impossible.

In 2019, I changed paths and began working as an English–Italian translator. I studied and worked as a freelancer, collaborating with several agencies and clients for years—though none ever offered full-time employment (I know it's hard to find a full time contract freelancing, but after 5 years?). Still, I was happy, my clients were satisfied with my work. I earned more than enough to get by in Italy and enjoyed the work.

Back in 2019, tools like Google Translate were widely mocked in the translation community. We could easily spot machine-translated text, and we felt confident that no machine could truly replace us.

But something changed around 2022–2023. Large Language Models started producing output that was “good enough” to fool non-specialists, and good enough for large volume-low-quality jobs (like translating UI/UX, web marketing content, low-tier advertising and articles). I began getting complaints from clients who had unknowingly purchased machine-translated content. At first, this led to more work for me, as I was hired to fix these flawed translations.

Then came ChatGPT and a visible shift in the industry. Starting in 2023, we saw a massive drop in demand—probably over 70% from my personal experience looking at job offers online and by clients. Companies began using AI to translate everything: websites, terms of service, contracts, blogs, and internal documents. The amount of work available shrank dramatically, but the number of translators stayed the same. Universities still churn out thousands of new language professionals every year.

Back in 2019, companies would reach out to me, asking me to work with them. Today, I’m the one reaching out—and often being ignored. Rates have collapsed. Where I used to earn $0.03–$0.05 per word (already considered low by industry veterans), now most offers are for post-editing machine translations at around $0.01 per word. The more advanced the AI is for a language pair—like English, Italian, French, German, Spanish, or Portuguese—the lower the pay. Clients don't care if it takes me 2 hours to go through a text and proofread it. The AI takes 30 seconds to write it, so they want the translators to proofread it in 5 minutes!

Now, at 32, I find myself forced to start over once again—searching for a new career from scratch. But nearly every job posting related to soft skills is either “entry-level” with two years of experience required, or a ghost listing that never leads to a response. Just this week, I was interviewed for an internship—and rejected—by an AI after completing an automated test.

I understand that I’m probably a mediocre worker. I’m not part of the top 5%, and I’ll likely never learn to code or network my way into a company like Google. But what does the future hold for people like me? For the other 95% of the population who can’t afford to constantly reskill or upskill every couple of years just to keep up?

-Anonymous

Those who turned English translations of Chinese into other languages were the first to go

I’ve been working in localization for a Chinese game company for a number of years. I enjoy my job—the vast gulf between Chinese and English means I have a lot of creative freedom to make tweaks and changes to the text and add things like little references for English speaking audiences to enjoy. But over the past few years there’s been an increasing switch toward using AI for translation work.

We’ve so far managed to convince management that Chinese-English translation should remain human. But we also do what’s called pivot translation—that is, translating from Chinese to English then English to French/Spanish, etc. In this field we generally used skilled freelancers, but now the shift is to MTPE. I hate the fact we’re taking money away from skilled translators, but we’ve had no way to push back on it as the cost savings have been significant, those players don’t seem to mind, and the higher-ups don’t seem to care much about the opinion of players from those language groups.

-Anonymous

Salaried translators were given a choice: Take a 50% pay cut, or resign

I currently work at a company focused on localizing adult games from Japanese to English. (Yeah, hentai games.) I used to be one of the top 3 members of said company until this April, and I've been with the company for longer than both of the other two managers.

The company I work for has been actively avoiding the use of AI in our translations due to concerns over the final output's quality at every level of our localization process. (I, myself, was one of the translators within it advocating against the use of AI.) However, this has not been true for the company's competition. In recent years a domestic Japanese publisher of these games has decided to enter the English localization market, and they have had no qualms against using AI in order to churn out mediocre products faster and at greater scale, publishing the slop on Steam.

As a result, the company I've been working for has begun struggling to acquire licenses to titles to work on period. Because of this, our board of investors chose to divest and sell off the company to the man who was its acting general manager at the time. Then he, effective this April (the start of this financial year), came to all of us who were in any kind of salaried position and told us we could take a 50% (or higher) pay cut to stay on, or we could walk.

He gave us all claims that it was to “optimize efficiency” or “refocus on more profitable performance tasks,” but away from the others he admitted to me that it's mostly so the company could have more free capital on hand in order to compete for licenses better.

So in short, we've all had our salaries slashed because our competitors are unafraid to make liberal use of AI to churn out barely-passable slop translations of adult titles so they can flood the market and monopolize the supply-side (the original developers).

The worst part of it all? We're not even seeing much outcry or antipathy from the fanbase, which is usually quick to criticize localizations. So we're kind of left to conclude that either the developers don't care that their titles are only seeing middling sales abroad, or customers don't care if their porn is using AI slop so they're willing to buy it anyway.

-Anonymous

AI didn’t even improve efficiency; it just made the work worse

I'm a freelance translator and interpreter. (interpreters do what people call “live translation”) and I've loved this job for as long as I've had it. I started translating when I was fifteen (helped family members with their jobs) and have been interpreting for the last five. Now, my language pair is a common one, so my job wasn't JUST killed by AI, but it sure as hell didn't help. English is the lingua franca of our day (fun fact, lingua franca comes from French being the lingua franca of ITS day), so my job as an interpreter was going the way of the dodo sooner or later and I knew it; I didn't expect it to die off this soon, but them’s the breaks.

By 2023, ChatGPT and DeepL had burst into the scene and, suddenly, no more translations (except some legal and sworn texts, which I like); which, in a vacuum, ok. I mean, not great, I'm not getting paid, but whatever. My issue is not (only) that I wasn't getting any money, but the final product itself: the translations these things offer are, in the best of cases, ok. Now, if you've never translated, you might think this job is just going over a text with a dictionary and taking Spanish word A and plugging English word A where it was (that's my pair). I haven't worked with any other languages professionally, but I can guarantee that it's not how it works for me: languages have nuances that are painfully obvious if you use them (you might not be consciously aware of them, though) but, if you don't, are invisible. Most people don't know this, so what we're getting now are mediocre translations that are a ghost of their originals and which lose everything that made them stand out. And, because people can't tell, they're happy with them.

[These services] don't even save you that much time. They just change your workflow. I'm a lazy person (I swear this makes sense) so I learned fast to hand in ok first drafts. This means I'd devote about 60 to 70% of my time to writing a passable draft and then the rest of the time just editing it into shape.

With DeepL, five minutes are just dumping the text in and out, and I spend at least a day working whatever it gives me into something resembling anything closely related to decent. Then, a day punching it into an ok shape, and the rest of the week just tweaking it into something I can be fine with delivering. The end product? Something okay. For legal stuff that's fine, for anything else, not really.

The thing is that people are okay with it and are offloading a lot of work onto this service and ChatGPT.

And that's really one of the biggest issues I have with AI and what it's doing: it trivializes everything and turns it into “content” that is “good enough,” turning everything into a worthless mush; just stuff to fill our (work)days. AI didn't kill translation, it didn't kill my job, it killed everyone's capacity to care about anything but the bottom line; because the people who have traded me in for DeepL aren't even keeping the money they would've paid me or taking time off; they're just being forced into different bullshit jobs while some C-suite goof is off golfing or whatever it is they do for “fun” all while they talk about efficiency and numbers.

-Anonymous

AI-happy execs don't appreciate how much of game translation is about nuance

I do work in translation, but my main income comes from legal transcription editing. AI makes it more fucking annoying for sure, even though I'm an editor and not a direct transcriber anymore. I have to clean up stupid AI mistakes constantly when just paying a real person to do this would have made it smoother on all ends. The AI used cannot even determine the difference between the word stenography (a word that comes up a lot, since these are court proceedings with court reporters and videographers present) and sonography.

In the case of translation, what I really have experience with are the opinions of people in niche game communities who want to play older and/or untranslated Japanese games. I worked on the newest English translation of a re-release of a vaguely popular game, which thankfully did not use any AI. A lot of people seem to understand that machine translated or AI translated games are not going to give you an accurate or enjoyable experience, but a growing number of people seem to think that they're fine and that "any translation is better than no translation."

These people don't seem to appreciate how much of game translation is about nuance, tone, and characterization. An AI is not going to be able to accurately translate puns or preserve the rhyming scheme of a song while keeping the translation accurate as well. Translation is not just about looking up words in a dictionary and pasting them into a document. Hell, it's rare to even find a word or phrase that has one single translation that can't be interpreted to mean something slightly different. A good translator needs to be not only knowledgeable, but flexible and creative as well. So much goes into this line of work, but it's rare that people fully recognize the full extent of effort it takes to produce it.

-Anonymous

AI systems aren’t just driving down wages, they’re flattening culture

I've been a freelance French-to-English translator in Quebec for 15+ years… From 2020 to 2023, I was so busy that I was turning down work, and still easily clearing six figures, primarily from freelancing for a financial institution (FI) that was paying $0.25 per word.

In hindsight, there was a brief period when translators were able to leverage tech (in my case, CAT tools) to their advantage, but as soon as AI started blowing up in the media, the secret was out. In 2024, the FI restructured its department, hiring more in-house translators with what seemed to be the goal of doing as much MT post-editing (MTPE; industry lingo for going through a machine translation line by line and making sure there are no mistakes) in-house as possible and reducing outsourcing. I chose not to apply for an in-house position because I was not interested in working as an employee after having been self-employed for my entire working life (I also saw the writing on the wall and knew that the work would mostly be MTPE for in-house translators). It wasn't long before I stopped receiving freelance work from the FI. I also chose not to pursue other freelance or agency work in MTPE because it is mind-numbingly boring, frustrating, and not worth the lower rates, so I can't speak to what the agencies have been offering their freelancers.

In 2024, my income went down 60%, and this year it's looking like it will be 80% lower than between 2020 and 2023. Of my contacts in the field, many are pursuing other careers and/or have left the profession altogether. I did pursue training in another (artistic, much less lucrative) field when I was younger, and I plan on pursuing that path, because this industry is just depressing the hell out of me. Thankfully I live in a place with a strong safety net (universal healthcare, subsidized childcare, child benefit payments), I have a partner earning enough, we have enough savings, and we own our home. If I was in a different position, I think I'd likely have to start from scratch or go to school, because there really aren't many transferable skills that are safe from AI (think copywriting, editing, etc.).

While I do think that some AIs are decent at translating, MT will need human intervention for the foreseeable future. But no translator will ever tell you they got into this field to do MTPE.

More than anything, though, I find it disheartening that instead of a society that once valorized translators as intercultural communicators and professionals who could uphold a bilingual society, we're flattening culture with AI systems that don't allow for a more organic exchange between languages. Quebec, in particular, has a rich linguistic landscape in both French and English, which can be owed to the cross-pollination of languages and cultures through human interactions, one of which is/was translation. Also, it just sucks that capitalism has found another way to undermine workers.

I was happy to have what I perceived to be the power to be on my own and work according to my wants and needs. But that option is no longer open to me.

-Laura Schultz