In short
- No US federal law currently says you have to disclose AI-generated content.
- A federal bill exists (introduced June 2026) but has not passed. A near-identical attempt in 2023 failed.
- California's SB 942 is a real law, in effect since 2 August 2026. It applies to the providers of AI tools, not to people who post with those tools.
- US and EU law differ in who they apply to, not in how strict they are: the deployer in the EU, the tool-maker in California.
- Platform rules (Fanvue, Instagram, TikTok) apply whatever any government says. This site covers them separately.
On this page
Mostly no. Today, no federal law says that someone who posts AI-generated content has to disclose it. One state law is already in force. It applies to the company that built the AI tool, not to the person who posts. The EU rule works the other way round. If part of your audience is in the EU, EU compliance for AI personas is the other half of this question.
No US federal law says you have to label AI content
Senators Brian Schatz, John Curtis and Mark Warner introduced the AI Labeling Act of 2026 on 25 June 2026, which would require generative AI providers to attach a visible disclosure and a machine-readable record of the system used, and require large platforms to flag such content rather than strip disclosures. The article confirms the bill was first introduced in 2023 by Schatz and a different co-sponsor, and did not become law then either.
Music Business Worldwide — US senators revive bill to force AI content labels (opens in a new tab)checked
So this is a bill, not a law. It was introduced, and there is no sign that it has passed committee. The same idea was tried before and stalled. If you search for “US AI disclosure law”, you find marketing blogs that name dollar fines and start dates as if they were certain. Nothing in this bill backs them up. If a page quotes a specific fine, treat it as describing something that has not happened yet.
What the sources do not say
Verifying a claimed FTC enforcement rule with a specific per-violation penalty for undisclosed AI content
Several search results described a signed federal rule that was already in effect, with specific numbers, a specific effective date and FTC enforcement. The rule was said to cover disclosing AI-generated advertising content. None of those pages was a government source, a bill text or a news outlet. All of them were SEO-style marketing blogs, and several seemed to be written to promote AI-disclosure compliance tools.
Congress.gov and the bill sponsor’s own Senate page both returned errors instead of content when fetched, so the bill text could not be read directly either. A dated, independent news report did confirm one thing: S.4915 was introduced in June 2026 and referred to committee. Being introduced is not the same as being a rule with penalties already in force. The penalty figures from those pages are not repeated here, because nothing found could confirm them against a rule that was really enacted.
How do I find out if this bill passes? Ask it in #question-thread. Someone there has usually had the same problem.
California’s SB 942 applies to AI tool makers, not to you
California’s SB 942 applies to a “Covered Provider” — defined as a person that creates, codes, or otherwise produces a generative AI system with over 1,000,000 monthly visitors or users, publicly accessible within California. It requires such providers to offer a free public detection tool, an optional visible disclosure feature, and embedded, machine-readable watermarking for AI-generated images, video and audio. The law places no direct compliance obligation on individual users or social media posters — the burden sits with the companies building the systems.
Jones Day — California enacts AI transparency law requiring disclosures for AI content (opens in a new tab)checked
SB 942 is a law that has passed, not a bill. It took effect on 2 August 2026, after an earlier delay to match its timing to the EU AI Act. It works the opposite way to the EU rule covered elsewhere on this site. EU AI Act Article 50 says the deployerDeployer Under the EU AI Act, a deployer is anyone who uses an AI system in their own activity. If you run an AI persona, that is you. Open the glossary has to disclose. That is whoever publishes the content. SB 942 gives the job to the provider. That is whoever built the generator.
Say you generate a persona’s images with a tool like Higgsfield or WaveSpeed and post the result. Because of California’s law, those tools might add a watermark and offer a detection page. The law does not tell you personally to file, disclose or label anything.
What that means if you post from the US
You post AI-generated persona content and want to know if US law says you have to disclose it. At the time of writing, no federal law does. The one state law with real force does not apply to you either, because it applies to the tool.
Two other sets of rules can still apply to you:
- The EU AI Act may still apply to you if any part of your audience is in the EU, wherever you operate from. EU compliance for AI personas is the guide for that question. Unlike US law, it does not depend on where you are.
- Platform rules apply without any government law. FanvueFanvue Fanvue is a subscription content platform that officially allows AI creators and labels them as AI. Open the glossary’s own AI-content policy, Instagram’s account-level label and TikTok’s per-video disclosure each apply by themselves. They are covered in what AI OFM is, Instagram’s AI-generated profile label and does TikTok allow AI influencers, in that order.
None of this is legal advice. After the dates on this page, a bill can pass, a state can make a new law, or a court can rule differently.
For the same question in other countries, are AI influencers legal puts the US answer next to the EU, the UK and India. India’s IT Rules have a page of their own, because India’s rules work through labels on the platforms. If you have any EU audience at all, read EU AI Act Article 50.
Sources for this article
The platform rules, laws and numbers above come from these documents. The date says when we last opened each one and checked it.
Questions people ask
Is there a US federal law requiring AI content disclosure?
No. The AI Labeling Act of 2026 (S.4915) would create one. When this was written, it had only been introduced, on 25 June 2026, and referred to committee. It has not passed. An earlier version from 2023, introduced by the same senator with different co-sponsors, never became law either.
Does California require AI content to be labelled?
Yes, but the rule is not for you. SB 942, the AI Transparency Act, took effect on 2 August 2026. It says "Covered Providers" have to offer disclosure tools and watermarking. Those are companies that build generative AI systems with over 1 million monthly California users. If you post content made with someone else's tool, this law does not directly tell you to do anything.
If I only operate in the US, do EU disclosure rules still apply to me?
That depends on where your audience is, not on where you are. Article 50 of the EU AI Act applies based on where the content is made available, not on where the operator is. This article does not answer that question. EU compliance for AI personas covers it.
Should I disclose anyway, even with no US law requiring it?
That is a business decision, and this article does not make it for you. One thing is clear: no law does not mean no platform rule. Fanvue, Instagram and TikTok all have their own labelling rules, and those apply whatever any government does.