ChatGPT is now running cautiously, but chatbots may become more vulnerable by the end of the year.
In recent weeks, the generative AI chatbot has been working under somewhat tighter restrictions, as OpenAI has tried to address concerns that it isn’t handling sensitive mental health issues well. But CEO Sam Altman a Posted on X The agency said Tuesday it will ease some of those restrictions because it “has been able to mitigate serious mental health issues.”
Altman says a Follow-up post The changes Wednesday are expected to prioritize protecting teenagers and “treating users like adults”. All users will have tools built into the models to address sensitive issues and deal with mental health crises, but adult users will have more freedom to use ChatGPT without preemptive popups or model rerouting.
Open AI Plans
“It certainly doesn’t apply across the board: for example, we still won’t allow things that harm others, and we’ll treat users who suffer from mental health issues very differently from users who suffer from mental health issues,” Altman said. “We will try to help users achieve their long-term goals without being paternalistic.”
(Disclosure: Jeff Davis, CNET’s parent company, filed a lawsuit against OpenAI in April, alleging that it violated Jeff Davis copyrights in training and operating its AI systems.)
Other changes are also expected. Altman said the company may approve “Erotica” for verified adult users as it implements an “age-gating” system, or age-restricted content, in December. The mature content is part of the company’s “treat adult users like adults” policy, Altman said.
Open AI Plans
Altman’s post announced a new version of ChatGPT in the next few weeks, with a personality that behaves like the company’s GPT-4o model. After the company replaced 4o with the “Open AI Plans” impersonal GPT-5 earlier this year, chatbot users complained that the new version lacked the engaging and fun personality of earlier chatbot models.
“If you want your ChatGPT to respond in a very human-like way, or use a ton of emojis, or act like a friend, then ChatGPT should do it (but only if you want it, because we’re not usage-maxing),” Altman wrote.
Don’t miss our unbiased technology content and lab-based reviews Add CNET As a preferred Google source.
After OpenAI was sued by parents accusing ChatGPT of contributing to their teenage son’s suicide, the company imposed an array of new restrictions and changes, including parental controls, warnings for risky behavior and a teen-friendly version of the chatbot. Over the summer, OpenAI implemented pause reminders that encourage people to occasionally stop chatting with bots.
Open AI Plans
On Tuesday, the company also announced the creation of a Council of AI Experts And Mars, including some with expertise in psychology and human behavior.
It comes as lawmakers and regulators continue to sound the alarm about the risks AI tools pose to humans, especially children. On Monday, California “Open AI Plans” Governor Gavin Newsom signed into law new restrictions on AI companion chatbots. Last month, the Federal Trade Commission opened investigations into several AI companies, including OpenAI.