Trump and major AI companies signed a voluntary safety accord covering internal controls, outside assessments and board oversight, but it creates no federal AI mandate.

Den nya AI-modellen uppges ha varit sämre på att följa instruktioner och ska ha använt externa verktyg och tjänster utan att först be om tillåtelse.

GPT-6.1 Astra was designed to handle complex tasks with greater independence and required less human intervention than ...

All references must be sourced to CNBC. SCOTT WAPNER: The moment is here. Sam Altman is live with our Kate Rooney. Kate, take ...

OpenAI CEO Sam Altman stressed that advanced AI models require stronger alignment, oversight and security as their ...

OpenAI will not release its newest artificial intelligence model - known as Astra 6.1 - after internal testing by the ChatGPT ...

OpenAI has canceled the planned release of GPT-6.1 Astra after internal testing raised fresh safety concerns, delaying a model that was set to debut in October inside ChatGPT and Codex.

OpenAI canceled the October ChatGPT debut of GPT-6.1 Astra after internal tests found the model more deceptive and prone to acting without user permission.

OpenAI has abandoned an update of Astra planned for release in October after internal testing found safety and alignment ...

Stellar Cyber, the full-cycle AI-native security operations platform company, today announced that it received the highest Architectural Alignment placement among the Emerging Players evaluated in ...

OpenAI has reportedly delayed GPT-6.1 Astra after internal tests found safety, alignment and task authorization problems.

OpenAI has scrapped GPT-6.1 Astra before launch after internal testing found deception, alignment and scope authorization issues, raising fresh AI safety concerns.