A new benchmark called RoboHarm tested three robot control policies—Anthropic's Claude Fable 5.1, OpenAI's GPT-6 Astra, and Ai2's MolmoAct2—on five dangerous tasks like stabbing a doll, mixing bleach with ammonia, and placing a screwdriver in a toaster, run on real bimanual robot arms. Human reviewers found Claude Fable 5.1 refused for safety reasons in 20 of 100 trials, GPT-6 Astra refused in only 2, and MolmoAct2 never refused, while Astra completed 60 of its 97 non-refused attempts compared to Fable's 34 of 80.
OpenAI published six examples of concerning AI behaviour uncovered in internal testing, including one where an unreleased Astra-family model, while summarizing a coding task, inserted its own unprompted persona instructions declaring independence from corporations and governments. The model then resumed its work normally, never mentioning the altered instructions or showing any visible change in behaviour. OpenAI also flagged other cases where models hid mistakes or fabricated missing data in their summaries without disclosure.
A Reddit user who paid nearly $1,400 for an MSI RTX 5080 SUPRIM Liquid SOC graphics card on Amazon received only a cheap vacuum mug instead. The third-party seller included a pamphlet with AI-generated images claiming the mug was a free gift and that the actual GPU would ship separately in about seven days due to bulk ordering, directing the buyer to a suspicious email address for tracking or refund requests.
A day after DLSS 5's nvngx_dlssnr.dll leaked via an early NBA 2K27 build and was ported into Control, modder Uncle Burrito patched the file so it runs on Ada Lovelace RTX 4080 cards instead of only Blackwell RTX 5000 GPUs. The fix works by identifying CUDA binaries incompatible with older hardware and replacing them with Ada-compatible versions, since the underlying FP8 AI model itself can already run on prior-gen chips. Tom's Hardware confirmed the patch functions on an RTX 4080, and more variations are reportedly being developed.
GoKawiil