California’s New AI Chatbot Disclosure Law: What It Means for You

California’s New AI Chatbot Disclosure Law: What It Means for You

California’s groundbreaking law requires AI chatbots to clearly disclose when they’re not human, plus extra safeguards for minors — here’s what you need to know.

I’m writing this today because the recent changes in California’s rules around AI chatbots are big, and they matter—not just in the U.S., but globally. With the rise of generative AI and chatbots, we’re entering a new era, and one key player here is ’s law requiring chatbots to clearly disclose that they’re not humans.

In this post I’ll break it down—what the law says, why it matters, how it might impact businesses/users, and what to watch next.

What the Law Says

Key Provisions

Here are the major parts of the law:

The law (, or SB 243) requires that when a “companion” or AI chatbot interacts with a user, it must issue a clear notification that the user is interacting with AI, not a human.
For minors in particular, the law adds extra requirements, such as providing reminders every few hours that the user is talking to a chatbot, and protocols relating to self-harm or suicide ideation.
Companies must create safety protocols: if the chatbot detects self-harm or suicidal thoughts, it has to refer the user to crisis resources.
The law is set to take effect on January 1, 2026.
The law is part of a broader package of reforms in California addressing AI, age verification, social-media warning labels, etc.

What it explicitly targets

“Companion” chatbots (ones that might mimic or assume a human-like presence) rather than purely functional bots.
Interactions with minors or potentially vulnerable users.
Situations where user safety, emotional wellbeing, or psychological risk is heightened.

What it doesn’t do (or at least not yet)

It doesn’t ban chatbots for minors outright. One proposed bill () that would have done so was vetoed.
It’s focused on disclosure and safety protocols, not full regulation of all AI models or decision-making systems.
The scope is currently California state law—other states (or countries) may differ.

Why It Matters

Transparency and user awareness

Let’s be honest: we’re increasingly interacting with chatbots and AI without always realizing it. This law forces companies to make it clear when we’re not talking to a human. That can help users adjust expectations and decisions accordingly.

Protecting minors and vulnerable users

There have been documented cases where children or teens form emotional attachments to chatbots, or where interactions go off the rails (e.g., encouraging self-harm). By requiring disclosures and safety protocols, California is positioning itself to tame some of those risks.

Precedent-setting regulation

California is often a forerunner in tech regulation—think data privacy, consumer protection, etc. This law could serve as a model for wider AI regulation (either in the U.S. or globally). If you’re building or deploying AI chatbots (or planning to), keeping an eye on California means you’re likely preparing for what others will do next.

Impacts for developers/businesses

If you deploy a chatbot in a way that reaches California users, you’ll need to update your UI/UX to include the disclosure.
You’ll need to review safety protocols, especially if minors are involved.
You might be facing liability and compliance costs, especially if this becomes standard across jurisdictions.

How This Affects You (or Could Affect)

For users

When you talk with a chatbot, you’ll now get a formal notice: “Hi — I’m an AI, not a human.” That changes the dynamic.
Especially for under-18 users, you’ll see recurring reminders and safeguards.
As a user you’re more empowered to know when you’re dealing with AI and make decisions accordingly (for example: trust, emotional investment, data sharing).

For content creators/website owners/bloggers (like me)

Since you asked me to write this as if I were the blogger, here’s how I see it: if I run a site and integrate a chatbot (for example, to assist users with queries about my blog topics), I now have to check:

Does my site reach California users? If yes → I need to show the disclosure.
Is the bot being used in a way that might affect minors? If yes → implement additional safety measures.
Am I using the bot in an emotionally-intensive shape (support, companionship, advice)? If yes → greater risk.
Internally: track what data the bot collects, monitor user complaints, ensure the bot doesn’t simulate a human without disclosure.

As a student (hi, I’m Hakim), writing about these topics means I also have to stay up-to-date because readers will ask “does this apply to me in India?” or “what about bots in WhatsApp groups?” So the law has ripple effects even globally.

For companies and chatbot builders

UI/UX has to include disclaimers at the start of interaction, and possibly recurring reminders.
Systems need to detect self-harm ideation or risky behavior, and have referral paths.
Age-verification / break-reminder features may need to be implemented.
Legal/compliance teams will have to map “if we reach California, what local law implies?”
Business models may need adjusting (e.g., fewer kids-focused “companion” bots).
Risk of lawsuits or enforcement if they fail to comply or mislead users.

What to Watch Next

Timeline & effectiveness

The law goes into effect January 1 2026. That gives chatbot providers a window to prepare.
Enforcement details (who monitors, what penalties) are still to be fully fleshed out.
Whether other states adopt similar laws or federal regulation emerges.

Global implications

Even if you’re outside California (for example India, where you live), if your service reaches users in California, you may need to comply.
Other jurisdictions may follow: the EU, UK, other US states are watching.
International companies will need to build with “compliance by design” for multiple regions.

Technological & ethical evolution

Disclosure is just one piece. We’ll likely see more focus on: bias, mental health impact, emotional dependence, deep-fake interactions.
AI chatbots may need to be certified or audited in future.
Companies might differentiate: “human-plus-AI” vs “AI only” experiences, to comply with regulation and maintain user trust.

My Key Takeaways (What I’m Thinking)

Bold highlight: For the first time, we’re seeing law force AI-chatbot interactions to be transparent in everyday user situations.
Bold highlight: The focus on minors and emotional/psychological risk is significant — this isn’t just about disclosure, it’s about user protection.
Bold highlight: If I were developing a chatbot (or advising someone who was), I would treat California as leading edge. Better to assume similar rules will show up elsewhere.
As a user, I’m going to treat bots with a bit more caution: If I see “I am a bot” disclaimers, I’ll ask: “What else have they done behind the scenes?”
As a blogger (like me writing to you), I think this adds another layer of responsibility. When I talk about AI, I should mention: “And yes, you might be talking to AI, not a person.”

Frequently Asked Questions (FAQ)

Does this law apply worldwide?

No, it’s a law in California. But if a chatbot service is available to California residents (or reaches them), it may apply. And many companies will opt to implement globally to avoid regional fragmentation.

What counts as a “companion chatbot”?

While definitions vary, it generally means chatbots that mimic human-like interaction, social engagement, emotional support—not just purely transactional bots (like “check my balance” bot). The law picks up more when minors are involved or when the interaction could have psychological impact.

What are the penalties for non-compliance?

Details are still emerging. The law signals stiff requirements (such as clear disclosure, protocols) but enforcement mechanisms and fines are less clear yet. Usual consumer-protection/regulatory enforcement may apply.

Does this mean every bot must show “I am AI”?

Yes — at least for those under the scope of the law in California. The bot must inform the user that it’s not a human. For minors interacting with “companion” bots, reminders are required.

External Authority Links

Details on SB 243 from the governor’s office: [Governor’s Press Release]
Explanation of how chatbots and minors played into this law: [CalMatters article]
Legal insight into the original California bot-disclosure law (2019) for comparison: [TermsFeed overview]

Final Thoughts

To wrap up: this law marks a shift in how we’ll think about AI chatbots—not just as tools or tricks, but as entities that interact with humans and need to be treated with transparency and care. If you’re using or building chatbots—or simply consuming their output—you’ll want to keep an eye on: “Is this AI? Do I know it’s AI?” Because starting January 2026 in California, that will be mandatory in certain contexts.

AI Doodlescape

Search This Blog