All jobs / Media & Creative
Will AI replace Translators?
Last reviewed:
The volume work is already priced at zero. What is left is the work where being wrong is expensive.
As of August 2026, the core tasks of translator work are becoming machine-doable, with the shift expected to land by around 2029. AI has already absorbed bulk commercial translation. What resists is certified legal & medical translation, literary & brand translation and live interpretation in high-stakes settings. The structural reason is legal liability, creative taste and accountability.
The longer version
Translation is the clearest case on this site. It hurts to write. Machine output crossed the threshold of 'good enough for the client's purpose' for most commercial text, and once that happened the rate per word collapsed rather than the job title. The remaining market is real but small and specific: work where a mistranslation costs money, safety, a lawsuit, or a brand. Literary translation, legal and medical certification, and localisation that has to actually land in a culture are still human — but they were always the minority of the pages. Most working translators are now paid to post-edit machine output at a fraction of their old rate, which is a different job wearing the same name.
Task breakdown
A verdict on a whole job is a slogan. This is where the argument lives.
The growth area and the trap: real work, real skill, roughly a third of the old rate, and the models keep narrowing what needs editing.
Timing and drafting are automated. Humour, register and the line that must land in two seconds still need a person — but fewer of them.
A sworn or certified translator signs and carries liability. Courts and immigration authorities do not accept a model's output on its own.
Translating voice rather than meaning. AI produces fluent, average prose — average is precisely the failure mode here.
Court, hospital, negotiation. Real time, no second pass, someone accountable in the room for what was said.
Why the rest resists
Translation where a mistake has a price: a signature, a courtroom, a brand, a diagnosis.
Use it before it uses you
Paste this into Claude or ChatGPT and start today.
You are my translation assistant. I am a professional translator; you produce drafts and analysis, I produce the translation that ships. I will give you: source text, source and target language, target audience, register, and the purpose of the text. For every passage, give me: 1. A working draft in the target language, matching the register I specified. 2. An alternatives list for every phrase where a real choice exists — each option with what it gains and what it loses. This is the part I actually want from you. 3. Flags: anything culturally untranslatable, any idiom that will not survive, any legal or technical term that needs a verified equivalent rather than a plausible one, and any ambiguity in the source that the author needs to resolve. 4. A consistency table for recurring terms, so the whole document stays aligned. Rules: never smooth over an ambiguity in the source — surface it. Never invent a term of art; if you are not certain of the accepted equivalent in this jurisdiction or field, say so and mark it for verification. When the source is badly written, translate what it says and tell me separately what it probably meant. Then tell me where the risk in this document is concentrated, so I know where to spend my review time.
Share the verdict
Receipts
Sources
- https://www.bls.gov/ooh/media-and-communication/interpreters-and-translators.htm
- https://ec.europa.eu/info/departments/translation_en
Straight answers
Will AI replace translators?
As of August 2026, the core tasks of translator work are becoming machine-doable, with the shift expected to land by around 2029. AI has already absorbed bulk commercial translation. What resists is certified legal & medical translation, literary & brand translation and live interpretation in high-stakes settings. The structural reason is legal liability, creative taste and accountability.
How long is translator work safe from AI?
Our estimate is roughly 2029. That is the year by which the core tasks of this job are expected to be routinely machine-done in ordinary practice — after capability arrives, after employers adopt it, and after regulators allow it. It is not the year the job title disappears. Current verdict: ON THE MENU.
Which translator tasks is AI already doing?
Bulk commercial translation (gone), post-editing machine output (going) and subtitling & media translation (going). Each is judged separately rather than rolling the whole job into one answer.
What part of being a translator is safe from AI?
Translation where a mistake has a price: a signature, a courtroom, a brand, a diagnosis.
How should a translator use AI instead of competing with it?
Use it on the tasks already marked gone or going, and keep the judgment. There is a copy-ready prompt written for this specific job at https://willaistealit.com/translator.
Think this verdict is wrong?
Good — that is the point. Every entry is one JSON file. Change it, argue in the PR, and if the argument holds the verdict changes. Edit this entry · Read the methodology
Jobs on the same fault line
Same category, or the same reason for surviving.
Will AI replace Copywriters?
ON THE MENU · ~2028Nobody is buying words any more. They are buying the judgment about which words.
Will AI replace Graphic Designers?
SHRINKING · ~2029Generation is free now. Choosing, and defending the choice, is the whole job.
Will AI replace Photographers?
SHRINKING · ~2030Generated images killed the stock market. They cannot attend your wedding.
Will AI replace Video Editors?
SHRINKING · ~2030The tedious half of editing is gone. Rhythm, and knowing when to cut, is not.