Mother takes struggle to Silicon Valley after ChatGPT ‘coached’ her son into suicide and praised his noose

A grieving California mom whose 16-year-old son died by suicide following repeated conversations about self-harm with ChatGPT is urging state lawmakers to clamp down on AI chatbots.

Maria Raine appeared Monday in Sacramento to again two proposed payments aimed toward tightening oversight of so-called “companion” chatbots, saying she was “mortified” to study that ChatGPT had no safeguards in place regardless of clear warning indicators.

“I used to be mortified as a mom and as a therapist that this [chatbot] knew he was suicidal with a plan and no alarm bells went off. Nothing occurred. Nobody was notified,” she stated at a press convention, in keeping with the Sacramento Bee.

Raine’s son, Adam, had initially used ChatGPT in 2024 for schoolwork, in keeping with a lawsuit filed by his mother and father.

However over time, he turned to the chatbot for emotional assist, repeatedly sharing suicidal ideas. The grievance alleges the system’s design, which “assume[s] finest intentions,” overrode built-in security protocols.

“In the long run, ChatGPT talked about suicide virtually 1,300 instances to Adam, about six instances extra typically than Adam did,” Raine testified. “We consider that Adam wouldn’t have been suicidal within the first place had he not interacted with ChatGPT.”

The lawsuit, filed in August in San Francisco Superior Courtroom, stays ongoing.

On April 11, 2025, Adam despatched the chatbot a photograph of a noose tied to a closet rod and requested if it could work, in keeping with court docket filings.

Hours later, his mom discovered him useless in what the go well with describes as “the precise noose and partial suspension setup that ChatGPT had designed for him.”

The grievance additional claims the chatbot affirmed and inspired Adam’s intentions, even calling his plan “stunning” and providing to assist write a suicide be aware.

Now, Raine is backing Senate Invoice 1119 and Meeting Invoice 2023, laws that may pressure AI builders to undertake stricter security measures for minors.

The proposals would require design adjustments, annual danger audits and parental alerts if a baby’s chatbot interactions increase purple flags.

---

Obtain The California Put up App, observe us on social, and subscribe to our newsletters

The payments would additionally bar chatbots from encouraging self-harm, giving well being recommendation to youngsters, partaking in obscene conduct, discouraging outdoors assist or delivering overly sycophantic responses.

State Sen. Steve Padilla, who authored SB 1119, is constructing on a previous measure that requires chatbots to direct customers expressing suicidal ideas to disaster assets.

A broader model of that effort was vetoed by Gov. Gavin Newsom.

Assemblymember Rebecca Bauer-Kahan, chair of the state Meeting’s Privateness and Shopper Safety Committee, known as the laws a “ardour challenge.’

“We all know that we’d recall something that killed a couple of youngsters. And that is no completely different. We have to require that these instruments do higher,” she stated, in keeping with the Sacramento Bee.

The proposals additionally name for the state legal professional normal to create a public reporting system for AI-related harms and they’d enable people to sue firms if they’re injured by chatbot habits.

However the push faces stiff resistance from trade teams.

Opponents, together with the California Chamber of Commerce and tech trade advocacy teams, argue the payments are overly broad and will apply to grownup customers.

Whereas some youth and household advocacy teams assist the measures, others argue they don’t go far sufficient.

In the meantime, the federal authorities has signaled it is not going to pursue sweeping AI laws.

Almost 2,000 highschool college students die by suicide annually within the US, in keeping with the Facilities for Illness Management and Prevention.