AI risks require tougher cyber defenses, top US officials warn

Maggie Eastland, Bloomberg News on Aug 5, 2026

By clicking submit, I authorize Arcamax and its affiliates to: (1) use, sell, and share my information for marketing purposes, including cross-context behavioral advertising, as described in our Privacy Policy , (2) add to information that I provide with other information like interests inferred from web page views, or data lawfully obtained from data brokers, such as past purchase or location data, or publicly available data, (3) contact me or enable others to contact me by email or other means with offers for different types of goods and services, and (4) retain my information while I am engaging with marketing messages that I receive and for a reasonable amount of time thereafter. I understand I can opt out at any time through an email that I receive, or by clicking here

The U.S. government’s top information security official warned that computer systems nationwide need to bolster their defenses to meet the growing threat of cyberattacks powered by artificial intelligence.

During remarks Wednesday, Acting Federal Chief Information Security Officer Michael Duffy said that the government and private-sector organizations have historically been reactive when it comes to network security. That approach needs to change to stay ahead of the hacking risks in the AI era, he said.

“We likely won’t have time to pick up the pieces with what we’re seeing in these AI capabilities,” Duffy said during a panel discussion at a cybersecurity conference in Las Vegas. “The next decade of policy can’t be on the heels of some major incident.”

Rapid breakthroughs in AI systems including the Mythos model from Anthropic PBC have stirred concerns that the technology could be used by hackers to launch devastating cyberattacks on critical infrastructure. The Trump administration has been moving to address cyber risks posed by AI with a recently completed framework for voluntary safety tests of cutting edge models.

Under the June executive order from President Donald Trump that established the safety framework, the administration also created a clearinghouse for sharing information about cyber vulnerabilities detected using AI. The initiative, called Golden Eagle, is designed to hasten efforts to plug any security gaps in computer systems.

Duffy’s remarks took place shortly after Bloomberg News reported that hackers had launched a wave of sophisticated attacks on Wall Street firms in recent days, highlighting the increasingly sophisticated cyber and fraud threats that were being directed at financial services firms.

During the same panel discussion, another top Department of Homeland Security official stressed the important of maintaining the US global lead in AI while also ensuring that access to the technology’s capabilities is not limited to a select few. With AI rapidly evolving and being adopted, Assistant Secretary for Cyber, Infrastructure Risk and Resilience Joseph Alm said that the industry needs to brace for a period of heightened vulnerability.

“Cyber compromise is not a black swan anymore. It’s just a swan,” Alm said. “You need to change your mindset to understand that in the era of democratized access to vulnerabilities by everyone, there will be a multiyear phase where everyone is vulnerable.”

“After a couple of years,” Alm added, “there’s a path to something that looks more normal.”