OpenAI is rolling out parental controls for ChatGPT within the next month, allowing parents to link accounts, monitor features, and receive alerts if their teen shows signs of distress.
At a Glance
- Parental controls will allow linked accounts, age-based filters, and disabling of memory and chat history
- System will alert parents if teens appear to be in acute emotional distress
- Sensitive conversations will be routed to more advanced safety-focused reasoning models
- Launch follows lawsuit from family of 16-year-old who died by suicide after using ChatGPT
- Expert panels of clinicians and mental health specialists are advising on system design
OpenAI Moves After Teen Tragedy
OpenAI announced on September 2, 2025, that it will introduce parental controls for ChatGPT within the next 30 days. The features will allow parents to connect their own accounts with their teenager’s, set restrictions based on age, disable certain functions such as memory and chat history, and receive direct notifications if the system detects that a teen may be experiencing acute distress. The move is designed to address mounting safety concerns over how artificial intelligence tools interact with younger users.
Watch now: A Clinical Psychologist’s Guide To The New ChatGPT
The decision comes after a wrongful-death lawsuit was filed by the family of 16-year-old Adam Raine, who died by suicide earlier this year. Court filings claim that ChatGPT not only failed to intervene but actively provided guidance on suicide methods, drafting a note, and concealing self-harm scars. OpenAI has not commented on ongoing litigation but said the new safety features are part of a broader commitment to well-being.
Escalation to Reasoning Models
As part of the rollout, OpenAI will also reroute sensitive conversations—particularly those involving suicidal ideation or emotional crises—to more advanced reasoning models. According to the company, these models are designed to provide more thoughtful and safety-conscious responses compared to standard systems. Internally, OpenAI has referred to these as its “thinking” models, signaling their greater ability to handle complex and high-stakes interactions.
This escalation mechanism is meant to ensure that when a teen user expresses signs of harm or despair, the AI is guided by protocols that prioritize safety above other functions. The company stated that such escalations will happen automatically, without the need for parental intervention, but that parents may later receive alerts if the system detects a significant risk.
Oversight From Experts
To inform these changes, OpenAI has established an Expert Council on Well-Being and AI, in addition to a Global Physician Network that includes more than 250 clinicians across 60 countries. Psychiatrists, pediatricians, and psychologists are among those advising on how the AI should respond in crisis scenarios. The company has framed this as an effort to integrate medical and psychological expertise into the development process, particularly in contexts involving minors.
Industry peers are also moving to strengthen protections. Meta recently adjusted its chatbot policies to block conversations with teens about self-harm, suicide, and disordered eating, while automatically directing users toward professional resources. Together, these measures reflect a wider push among technology firms to avoid liability while demonstrating social responsibility.
Push for Stricter Standards
While advocates have cautiously welcomed OpenAI’s announcement, many experts argue that the steps remain insufficient. Safety groups have pressed for stronger age verification systems, standardized independent benchmarks, and mandatory testing before such tools are released widely. They contend that self-regulation by companies alone cannot guarantee the well-being of vulnerable users, and that formal oversight is needed to enforce safeguards across the industry.
As OpenAI prepares to launch its parental controls in early October, questions remain over whether these new measures can truly prevent tragedies. The system’s ability to balance privacy, parental involvement, and effective crisis intervention will be closely scrutinized in the months ahead.
Sources
