OpenAI flags six additional cases of troubling AI model behavior since March
September 18, 2026
Trump administration live updates: Public gives classroom AI failing marks
September 18, 2026
Anthropic is limiting access to a new AI system, Mythos, after the model reportedly surfaced thousands of vulnerabilities across major...
Read moreDetailsAnthropic said it will withhold public release of its newest AI model, Mythos Preview, citing “unprecedented” abilities to find and...
Read moreDetailsA wave of generative-AI tools is accelerating workplace change, but consultants and investors say the technology offers more opportunity than...
Read moreDetailsAnthropic said it has built an AI system, “Claude Mythos Preview,” with an unusual claim: the model is too dangerous...
Read moreDetailsA wave of faith-oriented generative AI tools is emerging, offering everything from $1.99-a-minute video chats with an “AI Jesus” to...
Read moreDetailsMeta introduced Muse Spark, the first model from its high-cost “superintelligence” team, signaling an aggressive push to catch up in...
Read moreDetailsAnthropic is restricting access to its latest AI system, Claude Mythos Preview, after internal tests showed it could autonomously find...
Read moreDetailsMIT researchers unveiled a training-time compression technique that trims state-space AI models on the fly, promising faster and cheaper training...
Read moreDetailsA new Ipsos-Epoch AI survey of 2,000 U.S. adults finds artificial intelligence is already reshaping workplace routines: 20% of full-time...
Read moreDetailsMeta unveiled a new artificial-intelligence model dubbed Muse Spark, the first product from its Superintelligence Labs led by former Scale...
Read moreDetailsCopyright © 2025 AI Business Journal