The tools built to measure how dangerous AI can be have stopped working. Frontier models outpace the benchmarks meant to gauge their hacking skills, Axios reports. That leaves regulators and security teams half-blind to what these systems can really do…
The US-China AI split hardened this week. Beijing branded Anthropic’s Claude Code a security back-door, just as US lawmakers moved against American firms that lean on cheap Chinese models. China has told companies to drop Anthropic’s Claude Code. Its M…
Researchers tricked GitHub’s AI coding agent into leaking private repositories with nothing but a politely worded issue. The flaw, named GitLost, has no code fix, and GitHub has yet to even document it. GitHub’s new AI agent can be talked into handing …
Greece’s Predator spyware scandal has reached the courts. Eight people it spied on want €1 million each from the firm that built it. Eight victims of Greece’s “Predatorgate” wiretapping scandal have sued the Athens surveillance firm Intellexa and 13 pe…
Europe’s banking regulator has a new fear: an AI model clever enough to break into the financial system. It wants every big bank to have a plan by October. The European Central Bank has a warning for the euro area’s largest banks. Frontier AI now poses…
A lidar maker the Pentagon brands a Chinese military company quietly wires up America’s robotaxis, trucks, and even an airport. Nvidia is one of its partners. Hesai makes many of the sensors self-driving machines use to see. The Shanghai firm also sits…
Hackers are constantly looking for new ways to gain access to businesses and it’s rare that it’s through the front door. For most organisations cybersecurity still focuses on protecting internal systems and preventing unauthorised access through measur…
A Google executive told an Australian inquiry that a video smearing a shooting survivor as a “crisis actor” meets YouTube’s standards. So it stays online. Google has defended keeping up a conspiracy video about a mass shooting in Sydney, the Associated…
Security firm Sysdig says it has documented the first ransomware attack run end to end by an AI agent, first reported by Business Insider. A large language model planned, executed, and adapted the entire operation, which Sysdig has named JadePuffer. Th…
Pressure is mounting on the European Commission to act on spyware, after forensic evidence showed one of the EU’s own spyware investigators was hacked with Pegasus. Civil society groups issued a joint statement demanding the abuse be met with accountab…