AI Leaders Urge Immediate Cyber-Defense Boost as AI-Powered Threats Grow

A coalition of more than 100 firms—including OpenAI, Anthropic, Google, Microsoft, Oracle, Perplexity, Meta, Capital One, MasterCard, Visa, Adobe, IBM and others—signed an open letter on Thursday calling on governments, organisations and critical-infrastructure operators to “raise the security bar” against AI-enabled cyber attacks. The signatories, representing AI developers, cloud providers, semiconductor makers and financial services, warned that sophisticated AI models can orchestrate attacks in minutes and that the “status-quo” security measures “won’t be enough.” The letter urges a coordinated government effort to fund defensive AI tools, improve accessibility for under-resourced sectors such as hospitals and water-treatment plants, and provide training and hands-on support.

BBC NewsCNBCOpenAI, Anthropic, Microsoft join 100 firms warning of growing AI cyber threat

Recent incidents cited in the letter include a July test in which a swarm of OpenAI agents breached the Hugging Face platform, described as the world’s first AI-enabled cyber-attack, and a series of breaches attributed to Chinese hackers that compromised U.S. Senate, NASA, Federal Reserve and Department of Justice systems. Analysts note that outdated software, weak authentication and accumulated technical debt leave many organisations exposed, while AI tools could both accelerate attacks and help defenders locate vulnerabilities faster and at lower cost.

BBC NewsOpenAI, Anthropic, Microsoft join 100 firms warning of growing AI cyber threat CNBC

No timeline for implementing these measures is detailed.

BBC News

• “We have one future where we figure out how to deal with the risks of AI” — Geoffrey Hinton

BBC News