Does an AI agent have to say it is an AI?
RavinaroIs disclosure legally required?
Increasingly, yes. The EU AI Act's transparency obligations require that people are told when they are interacting with an AI system, and those provisions apply from August 2026. Several other jurisdictions have moved in the same direction, and the trend is one-way.
This is general information rather than legal advice, and the exact obligation depends on where you and your callers are. But the practical answer is simple: build for disclosure now, because the alternative is rebuilding under a deadline.
What does a compliant opening sound like?
Something close to: "Hi, you've reached [business]. I'm an AI assistant, and this call is recorded so we can follow up accurately. How can I help?"
Two facts, one sentence, at the top of the call. Every agent we build identifies itself as an AI at the start of every call and in its first message on every channel. It is not configurable, and we will not remove it on request.
Doesn't telling people hurt conversion?
That is the fear, and it is not what happens. What annoys people is not being served by a machine — it is being served by a machine that was pretending, which they usually work out about ninety seconds in, at which point they are angry about the deception rather than the wait.
An agent that says what it is up front gets to be judged on whether it was useful. Anyone whose agent pretends to be human is selling you a future complaint.
What if a competitor doesn't disclose?
Then they are carrying a risk that has a date on it. The commercial argument for hiding it was always thin, and from August 2026 in the EU it stops being a business decision.
There is also a quieter cost. An undisclosed agent cannot be honest about its limits, so it cannot say "I don't know, let me get someone who does" without breaking character. That single sentence is what stops a bad call becoming a bad review.
Sources
The rules quoted above come from the primary sources below. They are the law itself, not a summary of it — worth reading before you rely on anything here. This page is general information, not legal advice.
- EU AI Act, Article 50 — transparency obligations — Regulation (EU) 2024/1689. Article 50(1) requires that a person interacting directly with an AI system is informed of that fact unless it is obvious; Article 50(5) requires the disclosure "at the latest at the time of the first interaction".