A White House AI task force wants federal regulators to go easier on organizations that report cyber incidents in good faith, according to Cybersecurity Dive's report ↗ on a statement the group issued after its second meeting on Nov. 12. It reported that the task force urged agencies to limit regulatory liability for voluntary disclosures of vulnerabilities or incidents for at least 12 months, with a possible extension. Inside AI Policy separately reported ↗ that a White House readout of the meeting cited the need for strong liability protections to encourage industry input on incidents and vulnerabilities. AINewsIQ has not been able to review the task force's statement or the White House readout, so the 12-month details and quoted wording below come from Cybersecurity Dive.
According to Cybersecurity Dive, the statement argued that AI has shifted the risk: organizations that "only a short time ago had reasonable protections in place are now more vulnerable," so prior enforcement standards "may be outdated." It is a recommendation, not a rule. The report quotes it as applying only where relief is "lawful, practical" and in the public interest, and the quoted language does not say which agencies or reporting regimes it covers, or how good faith would be judged. Semafor reported ↗, citing an administration official, that the task force is led by Jay Clayton, the director of national intelligence.
A crowded reporting calendar
Cybersecurity Dive also reported that the task force backed a long-term renewal of a 2015 threat-sharing law, the Cybersecurity Information Sharing Act, and CISA's pending incident-reporting rule. The law's liability protections for sharing threat data now run to Dec. 11, 2026, according to the Congressional Research Service ↗, and Nextgov/FCW reported ↗ that 35 industry groups want them extended through 2035. CISA's rule under the 2022 incident-reporting law missed its October 2025 statutory deadline, per the federal regulatory agenda ↗, and Nextgov/FCW also reported ↗ it has been under White House review since Oct. 1. The rule would require reporting substantial incidents within 72 hours and ransom payments within 24.
That leaves a tension. The relief, as reported, is aimed at voluntary reports, while the CISA rule is a mandate; CRS describes the two systems as meant to work in tandem. The quoted statement does not say whether a company filing a required report would qualify.
What the threat data shows
WIRED's Nov. 14 reporting ↗ tests the premise. Researchers who track ransomware leak sites agree that claimed incidents spiked in recent months, WIRED reported, with some counting nearly 3,000 claimed attacks over the summer, a 20 to 40 percent rise since spring by some estimates. These are criminals' own claims, which WIRED notes are often inflated, and AINewsIQ has not found these figures reported elsewhere. Cynthia Kaiser of Halcyon told WIRED the rise was mostly international and "in part fueled by AI."
Allan Liska of Recorded Future told WIRED that AI shows up in phishing emails and ransom negotiations, but "successful" groups remain "largely human operated." ReliaQuest's Brandon Tirado told WIRED that a wave of small new groups, some claiming dozens of victims within weeks, reflects AI making it easier to start a crew. WIRED concluded that fully automated ransomware has not arrived yet; Liska cautioned that could change quickly.
Our analysis
Our read: the evidence points to a shift in cost and access, not a leap in capability. AI appears to make convincing phishing cheaper and launching an operation easier, so more attackers try more often. More frequent and polished attacks can complicate a regulator's assessment of whether existing controls were reasonable. They do not eliminate an organization's responsibility to maintain and adapt its defenses. A rise in volume is different from defenses being outclassed by machine-run attacks, and the task force's language, as quoted, does not distinguish between the two.
Liska told WIRED that Recorded Future's attack count this year already exceeds all of 2025, but that attacks involving encryption "may actually decrease." WIRED reports that newer independent operators may skip encryption and rely on data theft and extortion. Data extortion raises the notification and data-protection questions regulators enforce, which is where the proposed liability relief could become especially relevant.
What to watch
- Whether Congress renews the 2015 law by Dec. 11, and for how long.
- Whether CISA's reporting rule clears White House review.
- Whether any regulator puts the reported 12-month relief into practice.
- Whether the official readout is published in full.
Ransomware trackers broadly support the task force's diagnosis, but their evidence describes AI making attacks cheaper and more common, not autonomous.
Evidence note: AINewsIQ could not obtain the task force's statement or the White House readout. Their wording here comes from Cybersecurity Dive's report, and the ransomware counts are criminal groups' own leak-site claims as reported by WIRED.
This article was researched and drafted with AI assistance, then fact-checked and edited by Paul Woll, who is responsible for its content. Sources are linked inline.


Join the discussion
Have a different perspective or a question about the evidence? Discussion is free with a verified email account. Every comment is reviewed before publication.
No published comments yet.
New commenters: 2 submissions per 24 hours, at least 5 minutes apart, until 3 are approved. Discussion guidelines and moderation →