You’ve sent over your files, discussed the deadline, and received a quote. Then your Language Service Provider (LSP) mentions AI. You might be happy about the potential savings, but you’d probably like to know what’s happening to your content before anyone gets started.
That’s a reasonable conversation to have. In Nimdzi’s 2025 industry survey, 82.4% of responding language service providers offered machine translation and post-editing. These tools are already part of everyday translation work.
Still, AI’s role in language services varies. We might use it to draft translations, suggest terminology, or support quality checks. Ask which tools will be used, what they’ll do, and where people will step in. “AI-assisted” alone doesn’t tell you enough.
Your product descriptions, employee safety instructions, and company slogan have different jobs to do. They shouldn’t automatically follow the same translation process just because they arrived in the same folder.
When we’re choosing between machine and human translation, we need to understand who’ll read your content and what could happen if something is misunderstood.
A customer browsing reviews has different needs from an employee learning how to operate equipment.
Agree which materials can use AI, which need a different approach, and whether those decisions vary by language.
If you have an internal policy that rules out certain tools or uses, share it before files enter the translation process.
Before you hand over customer information or an unreleased product announcement, you’ll want more detail than “we use a secure platform.” Your IT or security team may have questions, and it’s easier to answer them before the project starts.
The conversation about translation confidentiality should cover where files are processed and stored, who can access them, and when they’re deleted. Ask whether your content, prompts, or translations can be used to train or improve any models.
Put the agreed restrictions in writing, including how they apply to freelancers and technology suppliers. Also settle who will contact you if a data incident occurs. You shouldn’t have to chase that information during a problem.
You might hear “a human checks everything” and picture a specialist carefully comparing every sentence with the original. Your provider could mean someone samples a few passages. Those are very different services.
In Slator’s 2025 survey, 76% of the 336 translators answering its AI-use question reported using AI tools in their work. That tells us AI use is common among those respondents, but it doesn’t tell you how your project will be reviewed.
Agree what machine translation post-editing includes, who performs it, and what subject knowledge they need. Specify whether every sentence will be checked against the source and whether a separate reviewer is involved.
If your team has spent months agreeing on product names and brand language, we don’t want to discover that after delivery. Send approved examples, terminology, and anything you’d rather never see in your copy.
A shared understanding of your business’s terminology and voice gives reviewers something more useful than “make it sound natural.” Tell us whether you want a formal tone, everyday language, or something closer to how your sales team talks.
For a larger project, agree on a representative sample first. Include tricky passages and the languages you’re most concerned about. Then decide what counts as an acceptable result and who on your side has authority to approve it.
It’s understandable to expect savings when AI is involved. We should be able to explain where those savings come from and what work still needs to happen before you can use the translation.
When discussing translation costs, confirm whether the price includes human review, formatting, quality checks, and revisions. A fast first draft doesn’t necessarily mean your finished files will be ready that afternoon.
Also agree what happens if the AI output needs more work than expected. Who approves a change in approach, budget, or deadline? You should know before extra charges appear, with enough time to make a sensible decision.
By the end of the conversation, you should have a written project brief that captures these decisions. Good translation project management gives you a named contact, clear delivery dates, and a process for questions and corrections.
Include who owns the finished files and reusable language resources, who signs off, and what changes require your approval. We can then get started with the same expectations, and you can get back to running your business.
If you’re wondering whether your translation requirements could benefit from an AI-supported approach, we’d love to talk.
Consultations are free and there’s no obligation. You’re in safe hands with us as we’re ISO 17100 and ISO 9001 compliant, have over twenty years of professional translation experience, and have earned the trust of organizations around the world.