AI Cyberattacks: OpenAI, Anthropic Warn of Escalation

More than 100 technology companies, banks, and security vendors are warning that AI-enabled cyberattacks will become far more widespread within months. This comes from an open letter published by OpenAI Group PBC. The letter states there is a limited window to strengthen cyber defenses. Hospitals, water treatment plants, and internet infrastructure are identified as the assets most at risk. Major companies like Anthropic, Google, Microsoft, Amazon Web Services, and Cisco Systems all signed on. Financial institutions such as Capital One, Mastercard, Visa, and Citigroup also joined. The letter argues that current security measures will not be enough. It points to existing vulnerabilities that attackers exploit without AI. The signatories say AI models can give specialist skills to teams that otherwise couldn't afford them. Organizations are urged to make cyber defense a leadership priority, fix high-risk weaknesses, and improve standards for AI-generated code. Governments are asked to coordinate and fund protection for essential services. The U.S. National Security Agency and other federal agencies recently issued a similar warning. They reported that threat actors are using AI-generated exploitation scripts for reconnaissance against critical systems. This alert highlights an urgent need for collective action to protect vital infrastructure.