Breaking
Google Play Opens the Door to Third-Party App Stores, Starting Next Week ◆ Murati Knows OpenAI’s Secrets. Her New AI Signals She Prefers China’s. ◆ The White House Wants AI To Beat Hackers To The Patch With “Gold Eagle” ◆ The Remarkable Success Of CAR T For Lymphoma 10 Years Later ◆ Microsoft Warns 2 Zero-Days Already Exploited In Attacks: Update Now ◆ China Plans AI Agent Recalls. America Can’t Even Agree Who Regulates Them ◆ Google Faces Another Lawsuit Alleging Its AI Violated Copyrights ◆ China’s Mythos Moment Is Coming, But How Bad Is The Threat? ◆ Lindsey Graham’s Sudden Death Sparks Questions About Cardiac Arrest ◆ Tricking AI By Simply Switching Which Spoken Language You Use In Your Prompts ◆ Today's NYT Strands Hints, Answers and Help for July 12 #861 ◆ Obamacare Premium Increases Raise Questions About Affordability And ‘Death Spiral’ ◆
Dark mode
Light mode
ClareNow
Search
ClareNow
Toggle sidebar
AI → Neutral

Meta Adds Support and Parental Notification for AI Chats About Self-Harm

Amid a period of intense scrutiny over safety, Meta is beefing up its options for parental supervision and notification of their child's online activities.

CNET 2 min read 7/10
Meta Adds Support and Parental Notification for AI Chats About Self-Harm
Key Takeaways
  • Meta AI chatbot now detects self-harm, suicide, and eating disorder discussions in conversations with teens, triggering parental alerts.
  • Parental notifications sent via the Family Center — parents see a summary of the concern, not the full chat content.
  • Feature uses Crisis Text Line-trained machine learning models, tested with clinical psychologists for accuracy.
  • Over 20 US states have filed lawsuits accusing Meta of knowingly harming teen mental health — this feature aims to address some of those concerns.
  • Rollout begins in the US in August 2025, with international expansion planned by Q2 2026.
Meta is rolling out new parental notification features for its AI chatbot, alerting guardians when teens engage in conversations about self-harm or distressing topics. The move comes amid heightened scrutiny over social media safety, with regulators and advocacy groups pushing for stronger protections for minors. Meta's AI chatbot, available on Facebook, Instagram, and Messenger, will now automatically detect concerning dialogue about suicide, self-harm, and eating disorders, and send a notification to the teen's linked parent or guardian. The feature builds on existing supervision tools, including activity dashboards and time limits, but represents a significant step in proactive AI-driven safety. Parents will receive real-time alerts via the Family Center, but will not see the actual content of the chats—only that a concerning conversation occurred. Meta says the system uses machine learning models trained on crisis intervention data and has been tested with mental health professionals. The rollout follows a string of lawsuits alleging Meta's platforms harm teen mental health, and a Senate hearing where CEO Mark Zuckerberg apologized to families. Critics argue the notification system may deter teens from seeking help, while supporters see it as a necessary safeguard. Implementation is phased, starting in the US and expanding globally. This move places Meta at the center of the debate over AI's role in mental health and parental oversight.

"Our goal is to give parents better visibility while ensuring teens still have private, supportive conversations when they need help."

"We built this in partnership with mental health experts to balance safety and privacy."

"We are cautiously optimistic, but alert fatigue and potential teen distrust remain serious risks."

Frequently Asked Questions

Meta uses machine learning models trained on crisis intervention data from organizations like Crisis Text Line. The AI analyzes message content for keywords and patterns related to suicide, self-harm, and eating disorders. When a likely risk is detected, the system flags the conversation for parental notification.

Parents receive a notification via the Family Center that their teen may have been involved in a conversation about self-harm or distressing topics. They do not see the actual message content—only a general alert that a concerning discussion occurred. This is designed to balance teen privacy with safety.

Meta faces intense scrutiny from lawmakers, regulators, and lawsuits alleging its platforms harm teen mental health. The feature is part of broader efforts to improve child safety online and respond to demands for more proactive oversight, especially after Mark Zuckerberg's Senate testimony and multiple state lawsuits.

No system is perfect. Meta says the AI is designed to minimize false positives and false negatives but acknowledges edge cases. The tool is continuously improved with feedback from mental health experts. It currently only works in English and for users in the initial rollout countries.

Meta claims privacy is a priority. The AI scans messages in real time but only sends a generic alert to parents—not the chat text. Teens are not notified that their conversation was flagged. Some critics argue this could deter teens from seeking help, while others say the safety benefit outweighs privacy concerns.

The feature launched in the United States in August 2025. Meta plans to expand to other English-speaking countries by early 2026, with broader international coverage expected by mid-2026. Users must have their Family Center linked and supervision enabled.

Topics

Meta AI chatbot parental controls self-harm detection teen mental health social media safety machine learning Family Center

Original source

www.cnet.com

Read original

Discussion

Join the discussion

Sign in to post a comment or reply.

No comments yet. Be the first to share your thoughts!

Sign in
Enter your email to receive a one-time sign-in code. No password needed.
Email address