Mandate 72-hour reporting of critical AI safety incidents to a national authority

Second, companies developing the most capable systems should meet clear requirements, including independent audits, incident reporting, strong security standards, and whistleblower protections.

Secondly, the Government must mandate independent statutory testing for AI before its release, and a yellow card scheme that would be a national register where safety incidents, near misses and harms caused by AI were reported, tracked and acted on e...

With the rise of Artificial Intelligence in all aspects of our lives, coupled with the lack of oversight, we face a looming crisis. This legislation sets a strong precedent to ensure AI developers abide by important safety standards and transparency ...

There's no shared incident reporting infrastructure. So you'll have different incident monitors tracking failures independently and voluntarily, but there's no legal duty to report or to recognize—or any recognition between them.

Before 2027 dialogue, we should advance shared terminology, comparable risk classifications, consistent documentation, interoperable incident reporting, and multilingual evaluation methods.

In addition, significant AI safety incidents must be reported to state officials within 72 hours of discovery, and whistleblower protections must be honored for employees who raise legitimate safety concerns.

Developers must report critical safety incidents within 72 hours, or within 24 hours if the incident poses an imminent risk of death or serious physical harm.

Frontier developers are required to report such incidents to the Illinois Emergency Management Agency and Office of Homeland Security within 72 hours of discovery, or within 24 hours if the incident poses an imminent risk of death or serious physical...

The report also recommends mandatory reporting of serious AI safety incidents within about 72 hours.

The Artificial Intelligence Safety Measures Act establishes a framework requiring developers of the largest advanced AI systems to publicly disclose safety practices, report significant safety incidents, and maintain robust compliance processes.

Adopting a reporting window of about 72 hours would align with protocols already used in industries such as aviation, nuclear power, and public health emergencies.

FLARE-AI is the culmination of that effort: a one-stop shop that redirects your reports to the voluntary recipient orgs working with us, rather than scattering accountability across social media, journalism, bug bounties, and feedback forms. Read the...

I think it’s a really good initiative. I’m in support of anything that makes AI more transparent.

Until now, the AI community has lacked a formal pathway to report flaws to organizations equipped to coordinate a response, leaving AI systems vulnerable to failure or exploitation.

Adopting a reporting window of about 72 hours would align with protocols already used in industries such as aviation, nuclear power, and public health emergencies.

The serious frontier developers already run the evaluations, the red-teaming and the escalation drills. What they have never faced at the federal level is a legal obligation to tell the government, on the clock, when a model behaves dangerously.

As AI systems grow more and more capable, the government needs a reliable window into powerful AI models developing dangerous capabilities, like launching a cyberattack or helping build a biological weapon. Congressman Moran is leading a new vanguard...

The capabilities of advanced AI models are increasing by the day, and so are the risks of AI systems gaining the ability to self-improve, evade human oversight, and enabling attacks against our nation’s infrastructure. We need mechanisms to identify ...

America should lead the world in artificial intelligence, and that leadership starts with secure and reliable systems. This bipartisan bill creates a practical framework to identify and address vulnerabilities before they can be exploited, helping st...

As AI technologies continue to advance at an unprecedented pace, AI systems face growing threats that can expose sensitive data, create new cybersecurity vulnerabilities, and beyond. Congress cannot afford to fall behind in addressing these emerging ...

I’m proud to represent much of the Research Triangle Park, home to organizations and institutions that are driving innovation in artificial intelligence and cybersecurity. AI is already shaping how we work, communicate, and access information, and we...

Where the CSIRT receives notification of an incident under regulation 11, 11A, 12A or 14E which it considers to materially involve autonomous or adaptive systems based on machine learning, the CSIRT must share relevant technical information with the ...

The Under Secretary for Health oversees integration of AI-related risk monitoring into existing patient safety programs and ensures staff are trained to identify and report AI-related safety events.

Mandates that frontier AI contractors report security incidents (theft of model weights) to the DoD within 72 hours of discovery and/or report material vulnerabilities in the model or concerning behavior within 7 days of discovering the issue is mate...

Safety incidents in the four critical areas must be reported promptly.

Material incidents must be reported to the Board promptly, without waiting for the scheduled annual report.

Additionally, frontier developers would be required to report critical safety incidents to the Illinois Emergency Management Agency and the Office of the Attorney General within 72 hours, or within 24 hours if the incident poses an imminent risk of d...

Track, document and report relevant information about serious incidents and possible corrective measures to relevant regulators and/or the public in a reasonable timeframe. Reporting near-misses and corrective measures is good practice.

Establish a mechanism for companies to share information about incidents, misuse, and near-misses with a designated public authority. The system should emphasize learning and prevention over punishment, with appropriately scoped public disclosures th...

Constructive engagement should focus on three concrete needs: a clarification that AI-specific attack vectors including data poisoning, model manipulation, adversarial ML inputs, and supply chain compromise of model artifacts constitute qualifying cy...

A FRONTIER DEVELOPER SHALL REPORT ANY CRITICAL SAFETY INCIDENT PERTAINING TO ONE OR MORE OF ITS FRONTIER MODELS TO THE OFFICE WITHIN SEVENTY-TWO HOURS FROM A DETERMINATION THAT A CRITICAL SAFETY INCIDENT HAS OCCURRED OR WITHIN SEVENTY-TWO HOURS OF TH...

Timely reporting of critical safety incidents to the government is essential to ensure that public authorities are promptly informed of ongoing and emerging risks to public safety. This reporting enables the government to monitor effectively in the e...

Canada should establish mandatory incident reporting for serious AI-related events, fund technical research on how to detect and contain autonomous AI systems, and integrate AI risks into the emergency management structures already under review.

The Contractor or Service Provider must complete the CISA incident reporting form [...] and notify the Contracting Officer and any Government provided point of contacts as soon as possible but not longer than within 72 hours of the discovery of any c...

We propose a minimum of 72 hours for the preliminary report for high-risk systems, which would allow sufficient time to assess the facts and determine whether the event meets the definition of a “serious incident.”

Practices include: Integrating a clear process to address AI incidents; Establishing a pathway for staff and the public to report AI safety concerns.

Mandates that the Attorney General establish a formal reporting mechanism for safety incidents, which large chatbot providers must report within a 15-day window.

Proposed law requires a frontier developer to report any critical safety incident involving its frontier models to the department within 15 days of discovery. Further requires reporting within 24 hours if the incident poses an imminent risk of death,...

AI is a powerful engine of innovation, and I want to see it flourish, but not without accountability and not without human oversight. The rule of law should apply to this new frontier. This legislation ensures that when something goes wrong with a hi...

The Center for AI Standards and Innovation at the Department of Commerce should be codified and funded as the federal government’s central technical authority on AI systems, with an incident review function modeled on the National Transportation Safe...

A series of measures could significantly improve the current system: (i) clarification of the reporting duties, including the development of a common taxonomy of risk or harm, alongside the expansion of the persons and entities under duty; (ii) disco...

It recommends common safety standards, testing of frontier AI before deployment, incident reporting within 72 hours, and global coordination to prevent catastrophic cyber, biological and autonomous AI threats.

Government needs trusted mechanisms for model access, evaluation partnerships, and incident reporting that protect sensitive commercial and national-security information while giving agencies enough visibility to act.

A developer must disclose each safety incident affecting the artificial intelligence model to the attorney general within 72 hours after the developer learns of the incident.

A frontier developer shall report any critical safety incident pertaining to one or more of its frontier models to the office within seventy-two hours from a determination that a critical safety incident has occurred or within seventy-two hours of th...

AI systems are being developed entirely within private industry, with little visibility available to Congress or the executive branch. [...] Congress should set transparency requirements for developers of high-stakes AI systems, including requirement...

[It is] concerning that models with far-reaching impact are controlled by a private company. [...] Now would be a good time to agree on disclosure rules and oversight mechanisms.

The federal government needs a reporting threshold for all artificial intelligence companies that deliver services in Canada, where they must report to law enforcement, so there's no judgment calls in a back room that Canadians don't have a line of s...

Timelines for incident reports are often set with well-intentioned urgency, but if they implicitly assume that root-cause analysis or full impact assessment has already been conducted, they will either produce premature and unreliable reports or plac...

The same patchwork of state mandates that purports to make AI safer is guaranteeing we learn less about the systems being deployed, not more.

I appreciate the intent behind the amendments and the shadow Minister’s position of understanding but not supporting them, which I share.