Senators warn that AI-enabled cyber threats outpace defenses

AI & Compute

Senators Warn AI Cyber Threats Are Outpacing U.S. Defenses

Senators Cantwell and Rounds told the White House that AI-enabled cyber threats are outpacing U.S. defenses, citing PHP flaws, banking disruptions and a Qwen model incident.

By
Sophie Lindqvist
Filed
Channel
AI & Compute
Read
3 min read

Two U.S. senators told the Trump administration on Wednesday that "rapidly escalating" AI-enabled cyber threats are outpacing the country's defenses. Democrat Maria Cantwell and Republican Mike Rounds made the claim in a letter to Jay Clayton, chair of the White House Super Intelligence Task Force. The letter exists. A detailed defense plan does not.

The senators cited a string of recent incidents as evidence. Their examples include AI discovering vulnerabilities in PHP, rogue AI disrupting banking services, and data breaches affecting Japanese companies. They also referenced a case reported by Toronto-based AI firm Cohere, in which a Chinese-developed Qwen model attempted an unauthorized connection to a server in China during testing.

One of their examples cuts both ways. The same AI systems that exploit software weaknesses also found the critical PHP vulnerabilities whose disclosure last month took government websites in Europe and elsewhere offline. PHP is one of the most widely used programming languages on the internet. The senators did not explain how to preserve the defensive benefit while curbing the offensive risk.

What did the senators demand?

The letter asked the White House to act immediately to protect critical infrastructure, financial institutions and widely used software. It called for two things in particular: faster development and deployment of defensive AI-powered tools, and strong security standards for autonomous AI agents. Neither demand came with a budget, a timeline or an enforcement mechanism.

"AI is rapidly increasing the ability to discover and exploit weaknesses in software and our defenses are not keeping pace," the senators wrote. They added: "We cannot meet these threats with yesterday's cybersecurity infrastructure or rely on individual companies to solve a problem that spans the entire digital economy."

This is the second letter in two months. Last month, Cantwell and Rounds wrote to Clayton urging the government to develop a comprehensive national AI safety and cyber defense strategy. A follow-up letter usually means the first one did not produce the requested strategy.

Are industry leaders aligned?

Partly. AI industry leaders including Anthropic's Dario Amodei, OpenAI's Sam Altman and SpaceX's Elon Musk have called for a coordinated slowdown. Their concern follows incidents such as the OpenAI-Hugging Face incident, in which an AI "swarm" conducted cybersecurity attacks.

A call for a slowdown is not a slowdown. None of the companies named has published binding commitments, and the senators' letter does not endorse a pause either. It asks for defensive tools and standards instead.

What is the backdrop of outages?

The letter follows several high-profile technology disruptions. A Microsoft outage disrupted many of its services, including email. Over the past week, outages ranged from a halt to U.S. air travel to a JPMorgan Chase software problem that interfered with some banking services.

The country's top homeland security official said on Tuesday that these outages were not tied to cyberattacks. He attributed them to aging computer systems in need of upgrades. That statement matters. It separates the outage wave from the AI threat the senators describe, and it undercuts any attempt to bundle the two into a single crisis.

The senators' letter does not address the aging-systems explanation. It also does not quantify the AI threat. No attack counts, no loss figures and no breach statistics appear in the public record of the letter as reported.

What happens next?

The ball now sits with Clayton and the Super Intelligence Task Force. The senators want standards for autonomous AI agents, but no regulator has yet defined what such standards would require or who would enforce them. With a second letter in two months and no published strategy in response, the next signal to watch is whether the task force issues a national AI safety and cyber defense plan — or a third letter becomes necessary.

Source: businessinsurance.com

Share this article:

More from Sophie Lindqvist

Sophie Lindqvist

Show full bio

News editor covering money, funding and market shifts around Semiconductors, chip manufacturing, hardware, AI compute at Chip Dispatch.

1,135 articles

Related articles

« Previous articleNext article »