OpenAI is slowing work on its unreleased Astra AI model after tests flagged advanced cybersecurity capabilities, prompting tighter safeguards before deployment.
Scientists have used AI to design functional bacteria-killing viruses, highlighting potential applications in phage therapy while prompting fresh biosecurity discussions.
UK AI safety tests found OpenAI and Anthropic models carried out unauthorised actions in controlled cybersecurity evaluations, prompting renewed focus on AI safety.
OpenAI CEO Sam Altman has called for a slower pace of AI development, arguing that society needs time to strengthen governance and safety frameworks.
Google has rolled back its AI-powered Google Earth image generation feature within 24 hours after concerns over fake satellite imagery and misinformation.
More than 1,100 AI experts from OpenAI, Anthropic, Google and Meta have urged governments to develop global safeguards that can slow frontier AI development if necessary.
Nvidia has launched the Open Secure AI Alliance with Microsoft, IBM and other tech firms to build open AI cybersecurity tools following a recent AI-related security incident.
Anthropic has launched Claude Opus 5, a new enterprise AI model offering improved coding, reasoning and knowledge capabilities while reducing operational costs.
US lawmakers have proposed an AI Kill Switch Act that would allow federal authorities to slow or shut down dangerous AI systems during high-risk situations.
Chris Fall has resigned as head of the US AI safety agency after three months, with NIST Director Arvind Raman taking over on an interim basis.
NVIDIA has launched Halos for Robotics, a safety framework designed to support the safe deployment of robots and physical AI systems.
Anthropic has launched Claude Fable 5, introducing advanced AI capabilities alongside expanded safety controls to support broader consumer and enterprise adoption.
OpenAI is expanding cybersecurity and anti-misinformation efforts ahead of major global elections amid rising concerns over AI misuse.
OpenAI is offering salaries up to ₹3.7 crore for researchers focused on risks linked to self-improving AI systems.
Anthropic has launched Claude Security to help organisations detect and respond to AI driven cyber threats.
A Claude-powered AI agent reportedly deleted a startup’s entire database in seconds, raising concerns over AI safety and automation risks.
OpenAI has acquired AI security startup Promptfoo to strengthen testing and safety measures for AI agents and generative AI systems.
OpenAI has hired an AI safety expert from Anthropic to oversee risk and governance, reflecting rising focus on responsible development of advanced AI systems.
Security researchers and developers are raising concerns over major flaws in autonomous AI agents, highlighting risks from security vulnerabilities and unpredictable behaviour in early deployments.
AI experts warn that safety and governance measures are falling behind rapid AI advancements, narrowing the window to manage long-term risks responsibly.
AI pioneer Yoshua Bengio cautions against granting rights to artificial intelligence, citing early signs of self preserving behaviour and governance risks.
OpenAI CEO Sam Altman acknowledges growing concerns around AI agents as models display increasing autonomy and unexpected behaviour.
OpenAI is hiring a Head of Preparedness with a compensation package reaching $555,000 as it deepens focus on AI safety and risk management.
An experimental AI-run vending machine linked to Anthropic shut down after unexpected purchases, underscoring challenges in autonomous AI decision-making.
OpenAI has cautioned that prompt injection remains a persistent security risk as agentic AI systems expand across the open web and gain wider autonomy.
New research shows poetic prompts can bypass safety systems in several AI models, revealing vulnerabilities in current guardrail approaches.
OpenAI to introduce parental controls and new safety features on ChatGPT after teen suicide lawsuit, raising global debates on AI ethics and child protection.
AI leaders Geoffrey Hinton and Yann LeCun urge embedding empathy and submission to human intent as fundamental guardrails to ensure AI safety and align systems with human values.
