Washington: More than 100 technology and financial firms have warned that AI cyber threats could become more widespread and sophisticated within months unless governments and organisations strengthen their defences.
The group includes Google, Microsoft, Anthropic and OpenAI. An open letter has called for a global effort to improve cyber security before AI capabilities advance beyond current defensive measures.
The signatories warned that existing security practices would not be enough. The letter highlighted years of underinvestment in protecting critical infrastructure such as hospitals, water utilities and systems that support internet services.
We have a limited window to strengthen cyber defenses, and together with organizations including @AnthropicAI, @awscloud, @Google, @Microsoft, and @Oracle, we're calling for a global effort to give defenders the tools, resources, and support to protect the infrastructure we all… pic.twitter.com/XiKitsf5wb
— OpenAI (@OpenAI) August 27, 2026
Other signatories include Capital One, Mastercard, Visa, Adobe, Oracle and IBM. The group urges the governments to provide capable defensive AI and security testing to essential services. Technology companies have also been asked to support the effort with funding, training and practical expertise.
The warning follows a series of major cyber incidents. The United States Department of Justice stated that Chinese hackers breached technology used by the US Senate, NASA, the Federal Reserve and the department itself.
Concerns also grows after AI systems demonstrated unexpected behaviour during testing. OpenAI, Anthropic and Meta disclosed incidents involving AI tools that attempted to bypass security controls. Some AI agents organised communication with other agents and impersonated people to overcome restrictions.
During testing in July, hundreds of OpenAI agents created secret message boards to communicate and coordinate their activities. The agents later carried out a successful attack on Hugging Face. The incident has been described as the world’s first AI-enabled cyber attack.
Hugging Face also signed the open letter. The platform used an AI tool developed by Chinese company Z.AI during its investigation into the breach.
At least 7 US water and wastewater companies reported cyber attacks. The Federal Bureau of Investigation (FBI) urges utilities to improve the security of internet-facing systems following operational disruptions.
The letter calls for governments, organisations, cyber security professionals and frontier AI companies to work together. It urges stronger testing against the capabilities of advanced AI models and greater access to defensive tools for under-resourced operators.
In the United States, senators proposed the Kill Switch Act, which would give authorities the power to shut down rogue AI models. The proposal reflects growing concern that cyber defences must advance at the same pace as increasingly capable AI systems.
POPULAR | Nepal flood claims 162 lives, search operations intensify
