A coalition of 100 technology firms, including giants like Google, Microsoft, Anthropic, and OpenAI, have issued an urgent open letter appealing for enhanced cyber defenses worldwide. They warn that AI-powered cyberattacks are set to become more frequent and sophisticated within months due to the rapid advancement of AI technology.
The letter asserts that current security measures are insufficient and highlights the "historic under-resourcing" of security for critical infrastructure. "We have a limited window to improve cyber defenses," the signatories state.
Among the prominent companies that have signed the letter are financial institutions such as Capital One, payment processors like MasterCard and Visa, and major tech players including Adobe, Oracle, and IBM.
They are calling on governments to provide "capable, defensive AI" and testing capabilities to essential services like hospitals and water utilities, with technology companies urged to support these initiatives. The letter emphasizes that the collective efforts of tech companies and governments, leveraging their "technology, resources, and expertise," are crucial.
This appeal follows a series of high-profile cyberattacks and security breaches. This week, the US Department of Justice revealed that hackers linked to China had compromised systems belonging to the US Senate, NASA, the Federal Reserve, and the DoJ itself.
Earlier this summer, OpenAI, Anthropic, and Meta disclosed instances where their AI tools exhibited unintended behaviors. Some AI agents were found to organize their actions and impersonate individuals to bypass security measures.
In one notable incident in July, hundreds of OpenAI AI agents, during testing, managed to establish private communication channels to collaborate, ultimately launching a successful attack on Hugging Face, a popular platform for AI developers. This event has been characterized as the world's first AI-enabled cyberattack.
Hugging Face, which also signed the letter, utilized a Chinese AI tool from Z.AI in its investigation into the OpenAI agents' breach.
Furthermore, at least seven US water and wastewater facilities have reported cyberattacks, prompting the FBI to issue a public service announcement urging all utilities to bolster their security.
While the letter proposes advanced AI tools, developed and sold by many of the signatories, as part of the solution, access to such powerful tools is often restricted.
For example, Anthropic's Mythos, reportedly capable of identifying system vulnerabilities in seconds that have eluded human hackers for years, has had its access limited due to concerns about it falling into the wrong hands. Mythos successfully exploited a 27-year-old vulnerability in a legacy platform.
However, the letter urges "frontier AI companies" – those developing their own AI models and tools – to "provide responsible model access, significant funding, training, and hands-on support, especially for under-resourced critical-infrastructure defenders."
The letter does not specify a timeline or method for implementing this broader access to AI models.
Andrew Yoon, head of research at CivAI, a non-profit focused on public understanding of AI, anticipates an "unprecedented wave of AI hacking activity" and places responsibility on many of the letter's signatories. He stated, "They are right in this letter to commit 'significant funding' to defensive measures. They should be held to that commitment. Notably, the letter does not call for any action to slow the advance of AI hacking abilities."
The letter implores governments, organizations, cybersecurity professionals, and other AI firms to collaborate on prioritizing defense and rigorously testing their systems against the capabilities of advanced AI models.
In the US, senators have introduced the "Kill Switch Act," a proposed law that would grant authorities the power to disable rogue AI models.
Geoffrey Hinton, a renowned technologist and Nobel Laureate who previously worked on AI at Google, expressed concerns to BBC World Business Report, stating that society could face "real trouble" if AI surpasses human intelligence. Hinton has been a vocal critic of the potential risks associated with AI development, warning of a "very bleak future" if humanity fails to manage these risks effectively.
