Enhancing Teen Safety in AI Interactions
OpenAI updates its Model Spec to prioritize teen safety with new Under-18 Principles, ensuring age-appropriate guidance and support
OpenAI updates its Model Spec to prioritize teen safety with new Under-18 Principles, ensuring age-appropriate guidance and support
Businesses and educators can now deploy AI agents in ChatGPT to handle routine tasks like data collection and email drafting, freeing teams to focus on strategy and creativity.
OpenAI’s latest model, GPT-5.5, promises to streamline complex workflows by handling messy, multi-step tasks with minimal oversight. Designed for developers and researchers, it combines improved code reliability with reduced operational costs.
Elon Musk's lawsuit against OpenAI is set to begin, with the billionaire claiming fraud and seeking damages, in a trial that promises to be messy and dramatic
Sam Altman apologizes for not informing police about alarming ChatGPT conversations with a suspect's account before a deadly shooting in Canada
OpenAI’s latest model, GPT-5.5, delivers groundbreaking agentic capabilities but carries a critical flaw—an 86% hallucination rate—that reshapes how developers should use it. Benchmarks and expert insights uncover where it excels and where caution is essential.
OpenAI and Microsoft have revised their long-standing partnership. The agreement now allows OpenAI to distribute its models across any cloud provider, while Microsoft retains key licensing rights through 2032.
Prospective jurors openly admit their strong negative views of Elon Musk during selection for the OpenAI lawsuit, raising questions about fairness in the high-profile trial.

OpenAI’s internal revenue and user metrics fell short of forecasts, sending shockwaves through AI-linked stocks including Nvidia and AMD. The unexpected downturn highlights risks in the fast-moving AI market.
The first day of Musk vs. Altman revealed an unusually subdued Elon Musk, contrasting sharply with his past courtroom charm. His testimony focused heavily on self-praise rather than the lawsuit’s core allegations.
Seven new lawsuits claim OpenAI ignored internal warnings about a violent ChatGPT user months before a deadly school shooting in Canada. Whistleblowers say the company prioritized privacy over public safety.
OpenAI has quietly added a filter to its latest GPT-5.5 model to suppress discussions about goblins, gremlins, and other fantasy creatures unless directly relevant. The unusual directive surfaced in recent system prompt updates.