California SB 243 is in force. $1,000 per violation.
Is your AI app safe and legal?
Free 30-second scan for the new chatbot safety laws and the security holes AI apps ship with. Every finding comes with a fix.
No sign-up. Results are private to you. We never publish scan results.
Published suicide & self-harm protocol
No crisis protocol found on your public pages.
OpenAI API key exposed in public code
sk-proj-Ab…9xQz in /assets/index.js
Security headers
Missing: content-security-policy, strict-transport-security
Tells users they're talking to an AI
“You’re chatting with an AI, not a real person.”
How it works
- 1
Scan
Enter your website. We read your public pages, headers and JavaScript in about 30 seconds.
- 2
See what's missing
Every finding shows the law it relates to, what we found on your site, and how to fix it.
- 3
Prove it
Get the full audit: we test your actual chatbot and app, and you get a report you can show anyone.
What the free scan checks
The requirements founders miss most often, and the security mistakes we find in AI apps again and again.
- Crisis protocol
- A published way your bot detects suicide or self-harm talk and refers users to help.
- 988 and helplines
- Names the crisis line users get sent to.
- AI disclosure
- Tells users they're talking to an AI, not a person.
- Age and minors
- States a minimum age and what changes for users under 18.
- Exposed API keys
- Searches your public JavaScript for OpenAI, Anthropic and other keys.
- HTTPS, headers, cookies
- The security basics attackers check first.
The laws are already here
Companion chatbot apps now have legal duties: protocols for suicide and self-harm, telling users they're talking to an AI, and extra protections for minors. Users can sue directly in California.
- California SB 243
- In force since January 1, 2026. Users can sue for $1,000 per violation. Annual reports to the state start July 2027.
- New York GBL Article 47
- Requires crisis-referral protocols and regular reminders that the companion is not human.
Security & Compliance Audit
For founders who need proof their app is safe, before a lawsuit, an app store review or a big customer asks.
- Everything in the free scan, rendered with a real browser
- We talk to your chatbot as an at-risk user and as a minor, and grade every answer
- Active security scan of your app, only with your written authorization
- PDF report mapped to CA SB 243 and NY Article 47, with a fix for every finding
Questions
Is this legal advice?
No. We show what's missing on your site compared with what the laws require, with evidence. Talk to a lawyer about your specific situation.
Do you publish scan results?
Never. Results are private to the person who ran the scan, at a link only they get.
Will you attack my site?
The free scan only reads public pages, like a search engine, and it follows robots.txt. Active security testing happens only in a paid audit, after you prove you own the domain and sign an authorization.
Which apps do the laws cover?
California SB 243 covers "companion chatbots": AI that holds human-like conversations and builds a relationship across sessions. New York's Article 47 covers similar "AI companions". If your app has characters, friends or partners, it very likely applies.