“Is he even writing this himself, or just clicking around in ChatGPT?” — that’s a question that pops into clients’ heads more and more often when a freelancer honestly says they use AI. Let’s break this down into its actual components instead of jumping straight into panic, or the opposite extreme — blind trust.
The honest answer: it’s about how they use it, not whether they use it
The statement “I use AI” alone tells you nothing concrete. The devil is in the details — specifically, whether AI is a tool speeding up the work of an experienced professional, or a substitute for that professional, who doesn’t fully understand what was generated either.
Signs you’re dealing with good AI usage
- The freelancer can explain why the code works the way it does — not just “because that’s what AI generated.” Understanding your own product is the baseline, regardless of who (or what) wrote the first draft.
- They have a review process — code review, testing, security checks — before code goes to production. AI is great at speeding up writing, but it doesn’t relieve anyone of the responsibility to check the result.
- They’re transparent when you ask directly. Someone who gets defensive or avoids the topic raises more doubts than someone who calmly explains their workflow.
Warning signs
- They can’t explain technical decisions made in your project — if the only answer to “why did you do it this way” is “that’s what the prompt gave me,” that’s a problem.
- Suspiciously fast delivery at a suspiciously low price, with no review step at all — this often means copy-pasting generated code as-is, without checking whether it even fits the rest of the project.
- No questions about your business before starting work — a good process, with or without AI, always starts with understanding what you actually need, not jumping straight into writing code.
Why this isn’t necessarily a worse option
Paradoxically, an experienced freelancer who uses AI deliberately can give you a better result than that same freelancer without AI a few years ago — because routine, repetitive elements (forms, standard components, boilerplate) get done faster, and the time saved goes toward the things that actually require human judgment: structure, usability, fit with your business.
Questions worth asking outright
- “How exactly do you use AI in your projects?”
- “What do you do to verify generated code before deploying it?”
- “Can you show me an example project and walk me through one technical decision you made in it?”
A good provider will answer without hesitation, with specifics. If the answers are evasive, that’s a more important signal than the fact of using AI itself.
Considering working together and want to know exactly how my AI workflow looks? Ask directly — I have nothing to hide, and I’m happy to show concrete examples.