← Richard Brooks

Essay · May 2024 · The language industry

Collaborative AI agents and translation: what the TransAgents study found

21 May 2024 · updated November 2026 · 5 min read · 3 animated figures

In May 2024 I shared my thoughts on LinkedIn about the potential of AI agents and how they could be used to build self-managing AI systems. The case study I discussed looked at ways to monitor hundreds of news articles and pass the most relevant ones to a key account manager.

Around the same time I came across an academic paper that applied the same idea to one of the hardest jobs in translation: novels. “(Perhaps) Beyond Human Translation: Harnessing Multi-Agent Collaboration for Translating Ultra-Long Literary Texts”, by Minghao Wu, Longyue Wang and colleagues, built a team of AI agents that works like a translation company. The authors revised it in 2025 for the journal Transactions of the Association for Computational Linguistics.

It’s worth a close look, because the company it simulates looks a lot like a language service provider. Here’s what it does, what it found, what it got wrong and what it means, with three figures to explore.

What is TransAgents?

A multi-agent system for literary translation. Each agent is a large language model (GPT-4 Turbo in the study) given a role, and together they reproduce how a human translation company handles a book: a CEO, a Senior Editor, a Junior Editor, a Translator, a Localization Specialist and a Proofreader.

To make the simulation more realistic, the authors generated 30 virtual profiles for each role, complete with nationality, education, years of experience, specialisms and a rate per word. The work runs in two stages: preparation, where the team is assembled and a set of translation guidelines is written, and execution, where each chapter is translated, localised, proofread and given a final review.

Click any step to see what the agent does, and the job it mirrors in a human LSP.

Figure 1

A translation company made of agents

Preparation

Execution, for every chapter

The roles and stages follow Wu et al. (2024). The comparisons with a human language service provider are ours.

How do the agents work together?

The paper uses two patterns, and both will be familiar to anyone who has run a translation project.

Addition-by-Subtraction. One agent adds everything that might be relevant. A second removes whatever is redundant. They go back and forth until the second has nothing left to cut. The team uses it to build the glossary and the book summary.

Trilateral collaboration. One agent does the work, a second critiques it, and a third decides whether it needs another round. The authors found the third role critical to quality, because it stops the loop.

Figure 2

Two ways the agents work together

    The patterns are from the paper. The example exchanges are invented to show how each loop runs.

    How was it tested?

    On Chinese web novels translated into English: the 240 chapters in the test set of the WMT 2023 shared task on discourse-level literary translation. The authors compared TransAgents with GPT-4 on its own, a fine-tuned LLaMA model, and the human reference translation.

    They used three kinds of evaluation:

    On both preference measures, readers and the model preferred TransAgents’ translations over GPT-4’s and over the human reference. It did best in genres that need background knowledge, such as historical and cultural settings, and worse in contemporary ones.

    The low d-BLEU score and the high preference scores aren’t as contradictory as they look. d-BLEU rewards closeness to one reference translation, and TransAgents wrote more freely. Two experienced professional translators who reviewed a sample chapter described its style as the most expressive and engaging, and also noticed it sometimes left parts of the original out.

    What would it cost?

    The paper’s cost comparison is the headline most people remember. Using the American Translators Association’s suggested minimum of US$0.12 a word, a chapter of about 1,404 words would cost $168.48 to translate by hand. TransAgents translated the whole test set for about $500, or $2.08 a chapter: roughly 80 times cheaper.

    Try your own numbers, and see what happens when you add the human review the study suggests you’d need.

    Figure 3

    Agents, humans, or both?

    The agent cost uses the paper's figure of about $2.08 per 1,404-word chapter, at 2024 GPT-4 Turbo prices; model prices have changed since. The review rate is illustrative, not from the paper.

    What went wrong?

    The paper is honest about its weaknesses, and they matter as much as the results.

    What does it mean for language service providers?

    The striking thing about TransAgents is how closely it copies an LSP’s own workflow. Every role in the diagram is a role a translation company sells. Once those roles can be written as prompts, the question for an LSP becomes which parts of its process carry the value that clients will keep paying for.

    The study points at the answer. The agents were cheap, fluent and often preferred, and they still dropped content that nobody would notice without checking against the source. On this evidence, the parts that still need people are quality assurance, accountability, and judgement about what a client actually needs. Those are the services to build and sell around.

    When I first wrote about this paper I said I was sure the tool vendors were watching it very closely. For more on what this shift means for language companies, see Nobody asked the language industry, my comparison of LLMs for LSP sales, and YouTube’s live auto-dubbing. If you run a language company and want to work through what it means for yours, that’s the kind of work I do in localization consulting.

    Sources: Wu, M., Xu, J., Yuan, Y., Haffari, G., Wang, L., Luo, W. and Zhang, K. (2024, revised 2025) “(Perhaps) Beyond Human Translation: Harnessing Multi-Agent Collaboration for Translating Ultra-Long Literary Texts”, arXiv:2405.11804, to appear in Transactions of the Association for Computational Linguistics; American Translators Association, atanet.org.

    Pass it on: Share on LinkedIn

    Enjoyed this? Get the monthly letter

    Once a month: the best of the essays, plus my take on what changed in AI and the language industry.

    One email a month. You'll get a confirmation email first, and every letter has an unsubscribe link. How your email is handled.

    Read next

    YouTube is about to dub live streams. Who eats the error?

    From early 2027 YouTube will translate live streams as the creator talks. Your clients will ask for the same thing. The question is who pays when the machine gets a word wrong.

    October 2026 · 8 min read

    Nobody asked the language industry

    Sixty organisations signed up in New York to bring AI to 3.4 billion people in their own language. No language company signed, and the industry's fastest-growing business is about to get cheaper.

    September 2026 · 12 min read

    Richard Brooks writes on business, value, and how AI is changing both. He is the author of AI Strategy for Sales Teams. For work, there is my company. This essay is also available as markdown.