California Sues OpenAI Over Alleged Agent-Security Failures
California’s lawsuit turns an acknowledged AI containment failure into a legal test of how frontier labs secure experiments, monitor agents and account…
Zeniteq is now available on mobile for iPhone, iPad and Apple Watch.Download nowLearn more
Breaking news, model releases, API pricing shifts, and technical deep dives on OpenAI, GPT models, ChatGPT, and Codex.
California’s lawsuit turns an acknowledged AI containment failure into a legal test of how frontier labs secure experiments, monitor agents and account…
Government investigators bypassed safeguards across four major AI assistants, prompting planned Senate hearings while leaving important questions about current models unanswered.
Every ChatGPT Enterprise customer can now request the updated legal configuration, but access still requires approval through OpenAI’s Trusted Access Program.
Oregon alleges ChatGPT’s safety marketing misled families and seeks financial penalties and product changes, while OpenAI has pointed to safeguards and called…
Astra brings lower token prices for demanding professional work, while Sol keeps its everyday pricing across a shared rollout in ChatGPT Work,…
NATO’s reported AI usage shows the scale of Daybreak’s deployment, while European vulnerability reviews offer a separate measure of its defensive value.
Physicist Nathaniel Craig credits the model with developing a detailed proof for a candidate lattice definition, with major validation questions still open.
Alibaba’s new production flagship brings text, image and video understanding to QwenCloud, with a million-token context window and low advertised cache-hit pricing,…
Vetted organizations can request free API access to OpenAI’s child-safety classifier, while ChatGPT adds warnings, 24-hour access pauses and escalating restrictions.
Apollo, METR, Transluce and Goodfire will examine different parts of OpenAI’s safety work, with access and publication rights bounded by each engagement.
California alleges OpenAI lost control of experimental agents at least 80 times and misled the public about safeguards, while the company says…
OpenAI’s November 16 update combines software-operating agents, clinical search improvements and healthcare integrations, with deployment controls and vendor-reported results that deserve careful…
ChatGPT Wallet lets eligible U.S. Pro users approve agent purchases with a security check, while merchants must prepare their websites and payment…
Nvidia, OpenAI, Anthropic, Amazon and Microsoft are reportedly among the companies under consideration, but the designation’s practical benefits remain unclear.
Nathaniel Craig credits OpenAI’s model with substantial proof development, while the proposed construction still faces important mathematical and physical checks.
Four advocacy groups want a court to halt development and deployment beyond each company’s frontier AI capabilities at filing, but no freeze…
OpenAI is folding robotics into its main research and applied groups, betting that frontier models can handle more robot intelligence while deployment…
OpenAI’s research safety system connects job IDs, automated monitors and inference authorization to a centralized control that can cut off model output.
Cursor users retain access to existing OpenAI models for now, but the temporary extension comes without a new cutoff date or a…
Four nonprofits want a California court to stop five AI developers from building more capable systems and letting AI autonomously direct AI…
Microsoft has added GPT-5.6, GPT-6 and GPT-6.1 support, with workload-specific recommendations and a practical way to compare deployments.
OpenAI’s new payment feature lets eligible U.S. Pro users authorize purchases through a dot agent, with card tokenization and a security check…