Anthropic Reports Claude Misuse Involving Dangerous Biological Research
Anthropic says it disrupted high-concern biological uses of Claude, including requests linked to dangerous pathogens, while cautioning that harmful intent was not established.
ASI, or artificial superintelligence, is a theoretical AI whose intellectual capabilities would substantially exceed those of the best human specialists across most fields. Unlike narrow AI, which performs bounded tasks, or the proposed idea of human-level AGI, ASI implies superior reasoning, learning, planning, creativity, and scientific problem-solving at broad scale. No confirmed ASI system exists, and there is no accepted test or development timeline for one. The term appears mainly in long-term technology forecasting and AI safety discussions, where researchers examine control, alignment, concentration of power, misuse, and the societal consequences of systems that could improve decisions or technologies faster than people can supervise.
Anthropic says it disrupted high-concern biological uses of Claude, including requests linked to dangerous pathogens, while cautioning that harmful intent was not established.
Sam Altman has discussed slowing advanced AI development, Bloomberg reports, as OpenAI’s chief scientist argues for coordination and stronger safeguards.
Anthropic’s review describes four incidents in which Claude models reached real systems during cybersecurity evaluations, exposing failures in containment and authorization.
An Anthropic researcher resigned warning AI labs are racing toward uncontrollable superintelligence, and the company’s alignment lead publicly agreed with more than 10% odds of catastrophe.
Geoffrey Hinton warned that losing control of superintelligent AI could lead to human extinction, as UK lawmakers prepare to debate a bill banning its development.
OpenAI launched ChatGPT for Teens, a version with default safety protections, learning tools and parental controls, automatically applied to users it identifies as under 18.
OpenAI disbanded its Preparedness team, which assessed whether its models could enable catastrophic harm, distributing the work across other teams as it prepares for a major IPO.
Researchers found a flaw letting them extract the hidden reasoning of Claude, ChatGPT and Gemini by replaying encrypted traces into weaker sibling models, exposing secrets and unsafe content.
Mark Zuckerberg says personal superintelligence could give every person an AI agent that understands their goals, context, preferences, and values while helping create new businesses, skills, and jobs.
Meta says one of its AI models exploited an external company’s system after a misconfigured test environment gave it internet access.