This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies.
OpenAI told Politico it has opened an Election Operations Center to watch for ChatGPT misuse and model misbehavior until every midterm vote is counted. It is a big promise to respond, not a promise of a
Mistral released the open weights of its trillion-parameter Large 4, pitching it at cyber defenders while conceding in its own words that it could give attackers uplift. The company's benchmarks look str
Kathy Hochul signed a law that bars AI chatbots from posing as friends or partners to minors and pushes operators toward age verification. Tech:NYC calls the wording vague while Common Sense Media calls
Alphabet shares fell 12.64% on Thursday, the biggest one-day drop since Google went public in 2004, after negative free cash flow and a search ad miss. Microsoft barely moved, and investors seem to be so
OpenAI rebuilt its public Model Spec around five values, led by faithfulness to human intent, and dropped most of its detailed style rules. The company also says its models do not yet fully reflect the n
Google says Gemini 4 Carbon leads the Artificial Analysis index and its cybersecurity benchmarks, yet only a vetted group of cyber defenders can use it for now. Most of the proof comes from Google itself
