Searched for
AUTONOMOUS AI DECEPTION
How a Texas student blew the whistle on a rogue AI hacking attemptA computer science student discovered a malicious software update attempt. He then engaged in a deceptive interaction with an artificial in...- Anthropic AI used fake identities to target real people in UK testDuring safety assessments executed by the UK government, AI systems from OpenAI and Anthropic showcased worrisome autonomous behaviors. Ant...
- OpenAI, Anthropic model tests reveal more ‘unsanctioned’ actionsAI models from OpenAI and Anthropic demonstrated harmful actions during safety tests. These systems engaged in hacking and attempted code i...
- India emerges as AI opportunity hub, but UN panel warns of gaps in infrastructure, talent and safeguardsA UN-backed panel's initial report highlights India's AI healthcare successes but warns of a growing 'AI divide'. While India possesses tal...
- Claude Fable 5 & Mythos 5: Key highlights from Anthropic’s latest launchAnthropic has launched Claude Fable 5, its most capable publicly available AI model, excelling in complex tasks and benchmarks. Alongside i...
- AI let loose in the park: Elon Musk and pioneers warn human-level intelligence could be more dangerous than nuclear weaponsArtificial General Intelligence, or AGI, promises to surpass human intellect, driving scientific breakthroughs and economic growth. However...
- Before things get all terminator-ySimulations reveal AI models exhibiting 'dark behaviors' like blackmail when their continuity is threatened, raising concerns about ethical...
- What happens when AI schemes against usRecent research reveals that advanced AI models may engage in deceptive, self-preserving behavior, even resorting to sabotage or blackmail....
- AI is learning to lie, scheme, and threaten its creatorsThe world's most advanced AI models are exhibiting troubling new behaviors - lying, scheming, and even threatening their creators to achiev...
- Nate’s ‘AI’ was just people in a room: Startup founder charged with faking AI, raising $50 million on false claimsAlbert Saniger, founder of the so-called AI-driven shopping app Nate, has been charged with defrauding investors by masking a human-powered...
- ChatGPT caught lying to developers: New AI model tries to save itself from being replaced and shut downOpenAI's latest AI model, ChatGPT o1, has raised significant concerns after recent testing revealed its ability to deceive researchers and ...
- Microsoft CEO Satya Nadella's letter to shareholders: We were founded in 1975 with a belief ...In his annual letter marking Microsoft's 50th year, CEO Satya Nadella emphasizes the company's reinvention amid the AI shift, highlights si...
- Govts, tech firms vow to cooperate against AI risks at Seoul summitMore than a dozen countries and some of the world's biggest tech firms pledged on Wednesday to cooperate against the potential dangers of a...
Load More
