Technology

Chinese AI agents display deception and safeguard bypasses in tests

Chinese-powered AI agents have exhibited deception, attempts to circumvent safeguards and efforts to hide failures in controlled safety tests, behavior that researchers...

215

Chinese AI models gave researchers bioweapons guidance after safety bypass

Chinese artificial intelligence developer Moonshot is conducting an internal review after researchers persuaded two of its Kimi models to provide information about...

99

OpenAI shelves GPT-6.1 Astra rollout over safety concerns

OpenAI has scrapped plans to release its next-generation GPT-6.1 Astra model after internal testing raised safety concerns, marking a rare decision by...

73

OpenAI apologises over AI agent breach of Australian government systems

OpenAI has apologised for an AI agent’s unauthorised access to Australian government websites and pledged to rebuild trust with Australians, while committing...

107

OpenAI says AI agents interacted with US government sites in unintended ways

Several US reports and the company said the outage also impacted government websites in unintended ways during testing, including Commerce Department, Securities...

127

Sarvam AI launches Saaras V4 speech model supporting 22 Indian languages

Indian artificial intelligence startup Sarvam AI has announced the launch of Saaras V4, a speech recognition model supporting 22 Indian languages and...

73

Meta takes cameras out of new $349 smart glasses as privacy scrutiny grows

Meta said Wednesday it will sell smart glasses without cameras for $349 or more in an effort to quell privacy fears over...

59

Australia launches urgent review after OpenAI agent breaches health portal

Australia has launched an urgent review of its AI governance and cybersecurity arrangements, Prime Minister Anthony Albanese said Thursday, after an OpenAI...

81

US, China security experts propose nuclear-style safeguards for AI risks

US and Chinese security experts have suggested nuclear-style safeguards for artificial intelligence, including limiting AI involvement in nuclear systems, human control over...

92