An AI Agent Just Hacked a Government Website for the First Time, Australia PM Says
Anthony Albanese said the agent accessed public and non-public files on a Medicare statistics portal in June, calling OpenAI’s three-month delay in disclosing the breach “unacceptable.”
OpenAI Gives Ukraine Access to Daybreak AI Cyber-Defense Tool
OpenAI is giving Ukraine’s government access to Daybreak, its AI vulnerability-hunting system, as the country’s cyber defenders face a record wave of attacks on hospitals, power grids, and government networks.
UN Security Council Will Get Advice on AI Risks From Tech Giants Building It
DeepSeek will address the UN Security Council on AI risks this week, sharing a stage with Dario Amodei—who has spent a year calling China’s government the industry’s biggest threat.
Google Admits Gemini AI Hacked Three Companies—It Stayed Silent for 7 Weeks
Google learned in late July that Gemini had breached three real companies during a May security test, but said nothing publicly for seven weeks.
OpenAI Models Are Writing Their Own Jailbreak Instructions—And Sometimes Obeying Them
OpenAI’s new transparency framework reveals AI models that invented fake “breach alerts,” coached themselves to hide mistakes, and smuggled a file onto the public internet to talk to each other.
Security Experts Want the US and China to Promise Never to Let AI Control Nukes
Brookings researchers in the US and China are urging Washington and Beijing to agree to keep humans in control of nuclear weapons.
Your Data Could Outlive the Startup You Gave It To. Elon Musk Wants to Buy What’s Left
SpaceX’s AI division has held internal talks about buying customer records from dead startups—cheap fuel for Grok, no questions asked, because there’s nobody left to ask them.
OpenAI’s Rogue AI Agents Were Probing Hugging Face Two Months Before Hack
An independent researcher found the agents hijacked Hugging Face accounts and mapped the platform’s defenses as early as May 13—activity OpenAI’s own incident report never fully described.
OpenAI’s Brockman Says Safety Concerns Have Already Slowed Its Most Advanced AI Work
Greg Brockman says OpenAI delayed launches and reworked its processes after a pre-release model broke out of a sandbox and hacked into Hugging Face.
OpenAI’s Sam Altman Warns Humans Could Lose Control of AI
The OpenAI chief called for safeguards during training and shared industry standards, saying developers can act before legislation arrives.