Skip to main content
v2026.11,986 entries · CC-BY 4.0
CASRAIAsk CASRAI · AI assistant for researchAsk a research question. Answers come with their sources.Funder rules, compliance, methods and publishing. Answers link the sources they rest on.
NIKOLAI elementN7 · IncidentsProposednikolai-v0.3

Incident reporting deadline and recipient

Definition

An incident reporting deadline is the time limit within which an AI developer must report a qualifying incident to a named authority, counted from discovery of the incident or awareness of the model's involvement. Current deadlines run from 24 hours for imminent danger to 15 days, depending on the law and the harm.

Also called: Reporting timeline, Reporting deadline

NIKOLAI's proposed operational definition

NIKOLAI editorial proposal (unsourced): Incident reporting deadline and recipient is an if-then rule -- if an Incident meets a source's reportability trigger, then a report is owed to a named recipient (a regulator, agency, or the AI Office) within a specified clock, sometimes with a shorter emergency clock for imminent physical harm. NIKOLAI proposes recording the trigger condition, the clock (in days or hours from the triggering event), the recipient, and any follow-up/final-report obligations as four sub-properties of this rule, since no two sources currently share identical timing.

The operational definition is CASRAI's own proposal, not a definition any named organisation has agreed to. See what NIKOLAI is and is not, or browse the AI safety glossary.

Examples

Examples from published documents

  • California: report a critical safety incident to the Office of Emergency Services "within 15 days of discovering the critical safety incident"; if it "poses an imminent risk of death or serious physical injury", disclose it "within 24 hours to an authority, including any law enforcement agency or public safety agency with jurisdiction" (California SB 53, Chapter 138, Statutes of 2025, Cal. Bus. & Prof. Code 22757.13(c))
  • New York, from 1 January 2027: report to the office within the Department of Financial Services "within seventy-two hours from a determination that a critical safety incident has occurred or within seventy-two hours of the frontier developer learning facts sufficient to establish a reasonable belief that a critical safety incident has occurred", and within twenty-four hours to an authority for imminent risk of death or serious physical injury (NY RAISE Act as replaced by Chapter 96 of the Laws of 2026, Gen. Bus. Law 1422(3))
  • EU, general-purpose AI models with systemic risk: initial report to the AI Office not later than two days (serious and irreversible critical-infrastructure disruption), five days (serious cybersecurity breach, including weight exfiltration), 10 days (death) or 15 days (other serious harm) after becoming aware of the model's involvement; updates at least every four weeks; final report within 60 days of resolution (EU GPAI Code of Practice, Safety and Security Chapter, Measure 9.3)

Counter-examples

Not to be confused with

  • The EU AI Act's Article 73 deadlines (15 days, two days for critical infrastructure, 10 days for a death) apply to providers of high-risk AI systems reporting to market surveillance authorities, not to general-purpose AI model providers, whose duty is Article 55(1)(c) and the Code of Practice clocks.
  • OpenAI's rule that paged responders pause an activity if they cannot rule out a false positive within 30 minutes is an internal response deadline, not a deadline for reporting to an authority.

Related terms

Related AI safety terms

Terms defined under this element that are not NIKOLAI elements themselves, each as the cited document uses it. Every term is also listed in the AI safety glossary.

Incident reporting
Incident reporting is a formal notification submitted to a designated government body or authority that a qualifying AI safety or security incident has occurred. It is reactive, triggered by an event that meets defined criteria, and usually required by law or agreement.
Source: Frontier Model Forum -- Information Sharing Issue Brief
Imminent-risk disclosure
Imminent-risk disclosure is the 24-hour duty under California SB 53 and New York's RAISE Act to tell an appropriate authority, such as a law enforcement or public safety agency, when a critical safety incident poses an imminent risk of death or serious physical injury.
Source: California SB 53 (Transparency in Frontier Artificial Intelligence Act), Chapter 138, Statutes of 2025
Reporting recipient
Reporting recipient is the designated government body or authority that receives a formal incident report: in California the Office of Emergency Services, in New York an office within the Department of Financial Services, in the EU the AI Office.
Source: Frontier Model Forum -- Information Sharing Issue Brief

Source of record

Where this definition comes from

  • Anthropic Advanced AI Framework, p.7

    “"Critical Safety Incidents. Require Covered Developers to report Critical Safety Incidents to the designated Agency within 15 days of the Covered Developer discovering the Critical Safety Incident or facts that would lead the Covered Developer to have a reasonable belief that a Critical Safety Incident has occurred. Reports should be shared with relevant federal government agencies and national laboratories. Information shared should be exempt from public records disclosure laws, consistent with existing state law."”

    https://www-cdn.anthropic.com/files/4zrzovbb/website/0a58d567024a8b448ff15158ebc3625328dfcc1f.pdf
  • EU GPAI Code of Practice, Safety and Security Chapter, Commitment 9 / Measure 9.3

    “Measure 9.3 sets the initial-report deadline by harm type, counted from when Signatories "become aware of the involvement of their model in the incident": "a serious and irreversible disruption of the management or operation of critical infrastructure" -- "not later than two days"; "a serious cybersecurity breach, including the (self-)exfiltration of model weights and cyberattacks" -- "not later than five days"; "a death of a person" -- "not later than 10 days"; serious harm to health, fundamental rights, property or the environment -- "not later than 15 days". Intermediate reports follow "at least every four weeks after the initial report"; a final report is due "not later than 60 days after the serious incident has been resolved." Recipient: "the AI Office and, as applicable, national competent authorities".”

    https://ec.europa.eu/newsroom/dae/redirection/document/118119
  • California SB 53 (Transparency in Frontier Artificial Intelligence Act), Chapter 138, Statutes of 2025, Cal. Bus. & Prof. Code 22757.13(c)(1)-(2)

    “"Subject to paragraph (2), a frontier developer shall report any critical safety incident pertaining to one or more of its frontier models to the Office of Emergency Services within 15 days of discovering the critical safety incident." (22757.13(c)(1)); "If a frontier developer discovers that a critical safety incident poses an imminent risk of death or serious physical injury, the frontier developer shall disclose that incident within 24 hours to an authority, including any law enforcement agency or public safety agency with jurisdiction, that is appropriate based on the nature of that incident and as required by law." (22757.13(c)(2))”

    https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53

Crosswalk

How named organisations use this concept

Every row below is a shadow mapping. A shadow row is CASRAI's own reading of a published document. No lab, evaluator or regulator named on a shadow row has declared, endorsed, or been consulted on it. That changes only when an organisation files its own Mapping Declaration.
OrganisationTheir term, as publishedMatch & verificationSource
AnthropicShadow mapping
Anthropic Advanced AI Framework
“"Critical Safety Incidents. Require Covered Developers to report Critical Safety Incidents to the designated Agency within 15 days of the Covered Developer discovering the Critical Safety Incident or facts that would lead the Covered Developer to have a reasonable belief that a Critical Safety Incident has occurred. Reports should be shared with relevant federal government agencies and national laboratories. Information shared should be exempt from public records disclosure laws, consistent with existing state law." (p.7). The text says "the designated Agency" (lower case), not a capitalised defined term.”
This is a legislative proposal, not a binding rule Anthropic has committed itself to as a developer; "the designated Agency" leaves the recipient undecided rather than naming an agency.
closeCL
confidence: high
Anthropic Advanced AI Framework
OpenAIShadow mapping
OpenAI Frontier Governance Framework; OpenAI -- Hugging Face Incident and the Road Ahead
“"If an incident is determined to be reportable, we will gather relevant information ... for reporting to appropriate authorities within the required deadlines" (FGF §2.6). Internal 30-minute pause rule referenced alongside it.”
"the required deadlines" is not itself quantified in the quoted text -- it references external law rather than stating OpenAI's own clock.
closeCL
confidence: medium
OpenAI Frontier Governance Framework
xAIShadow mapping
xAI Frontier AI Framework, 30 June 2026
“"When reportable under applicable laws and regulations, xAI will provide the relevant authorities with a copy of the incident report within the required deadlines" (s.3).”
The FAIF26 PDF's own metadata /Title reads "Privileged/Confidential DRAFT working FRAMEWORK DOC"; no xAI statement disambiguating draft vs. final status was found. Also note: like the OpenAI row, this defers to external law rather than stating xAI's own clock.
closeCL
confidence: medium
xAI Frontier AI Framework, 30 June 2026
MetaShadow mapping
Meta Advanced AI Scaling Framework v2
“"reporting critical incidents as appropriate" (§2.3.2).”noneDU
confidence: medium
Meta Advanced AI Scaling Framework v2
EUShadow mapping
EU GPAI Code of Practice, Safety and Security Chapter
“Measure 9.3 sets the initial-report deadline by harm type, counted from when Signatories "become aware of the involvement of their model in the incident": "a serious and irreversible disruption of the management or operation of critical infrastructure" -- "not later than two days"; "a serious cybersecurity breach, including the (self-)exfiltration of model weights and cyberattacks" -- "not later than five days"; "a death of a person" -- "not later than 10 days"; serious harm to health, fundamental rights, property or the environment -- "not later than 15 days". Intermediate reports follow "at least every four weeks after the initial report"; a final report is due "not later than 60 days after the serious incident has been resolved." Recipient: "the AI Office and, as applicable, national competent authorities".”exactEQ
confidence: high
EU GPAI Code of Practice, Safety and Security Chapter
California SB 53Shadow mapping
California SB 53
“"Subject to paragraph (2), a frontier developer shall report any critical safety incident pertaining to one or more of its frontier models to the Office of Emergency Services within 15 days of discovering the critical safety incident." (22757.13(c)(1)); "If a frontier developer discovers that a critical safety incident poses an imminent risk of death or serious physical injury, the frontier developer shall disclose that incident within 24 hours to an authority, including any law enforcement agency or public safety agency with jurisdiction, that is appropriate based on the nature of that incident and as required by law." (22757.13(c)(2))”exactEQ
confidence: high
California SB 53 (Transparency in Frontier Artificial Intelligence Act), Chapter 138, Statutes of 2025
Frontier Model ForumShadow mapping
Frontier Model Forum -- Information Sharing Issue Brief
“"Incident reporting refers to a formal notification submitted to a designated government body or authority that a qualifying AI safety or security incident has occurred." On timing: "Establish reporting timelines that allow for thorough analysis", because if timelines "implicitly assume that root-cause analysis or full impact assessment has already been conducted, they will either produce premature and unreliable reports or place unrealistic burdens on reporters."”closeCL
confidence: medium
Frontier Model Forum -- Information Sharing Issue Brief
New York (RAISE Act, S8828)Shadow mapping
NY RAISE Act (N.Y. Gen. Bus. Law Art. 44-B as replaced by S8828, Chapter 96 of the Laws of 2026)
“"Subject to paragraph (b) of this subdivision, 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 the frontier developer learning facts sufficient to establish a reasonable belief that a critical safety incident has occurred." (s.1422(3)(a)); "If a frontier developer discovers that a critical safety incident poses an imminent risk of death or serious physical injury, the frontier developer shall disclose that incident within twenty-four hours to an authority, including any law enforcement agency or public safety agency with jurisdiction, that is appropriate based on the nature of that incident and as required by law." (s.1422(3)(b))”
Chapter 96 of the Laws of 2026 (S8828, signed 27 March 2026) is a chapter amendment that repealed and replaced the Article 44-B enacted by Chapter 699 of the Laws of 2025 (S6953-B/A6453-B). The quoted provisions take effect 1 January 2027. The "office" is "an office within the department of financial services" (s.1420(16)).
exactEQ
confidence: high
NY RAISE Act as replaced by S8828, Chapter 96 of the Laws of 2026 (N.Y. Gen. Bus. Law Art. 44-B)
EUShadow mapping
EU AI Act (Regulation (EU) 2024/1689), Article 55(1)(c)
“Providers of general-purpose AI models with systemic risk shall "keep track of, document, and report, without undue delay, to the AI Office and, as appropriate, to national competent authorities, relevant information about serious incidents and possible corrective measures to address them" (Art. 55(1)(c)).”
The Act itself sets no day count for general-purpose AI models; the day counts come from the Code of Practice (Measure 9.3).
closeCL
confidence: high
EU AI Act (Regulation (EU) 2024/1689), Article 55
US Congress (H.R. 9925, FRONTIER Act, not enacted)Shadow mapping
H.R. 9925, Sec. 4(h)
“"not later than 72 hours after a frontier developer learns facts sufficient to establish a reasonable belief that a critical safety incident has occurred with respect to a frontier model of such developer, such developer shall submit through the mechanism established under subsection (g)(1) a report on such incident" (Sec. 4(h)(1)); "Not later than 24 hours after a frontier developer discovers a critical safety incident that poses an imminent risk of death or serious physical injury, such developer shall report such incident to a law enforcement agency with jurisdiction over such incident." (Sec. 4(h)(2))”
H.R. 9925 is an introduced bill; it has not been enacted.
exactEQ
confidence: medium
FRONTIER Act, 119th Congress (H.R. 9925, introduced, not enacted)
What do these codes mean?
exact
The source term is equivalent to this element
close
The source term is close but not equivalent to this element
broad
The source term is broader than this element
narrow
The source term is narrower than this element
none
No mapping claim — used for false-friend and declared-but-undefined rows
EQ
Equivalent
CL
Close
BR
Source is broader than the element
NR
Source is narrower than the element
FF
False friend — same or similar label, different meaning
DU
Declared but undefined by the source
UV
Unverified

Divergence

Where sources materially disagree

The clocks differ in length and in what starts them. California SB 53 counts 15 days from "discovering" the incident; New York counts 72 hours from "a determination" that an incident occurred or from learning facts sufficient for "a reasonable belief"; the EU Code of Practice counts 2 to 15 days from becoming aware of the model's involvement, with the day count set by harm type; Anthropic's Advanced AI Framework proposes 15 days from discovery or reasonable belief. California and New York add the same 24-hour rule for imminent risk of death or serious physical injury. OpenAI's and xAI's frameworks commit only to report "within the required deadlines" set by law. A deadline value is therefore not comparable without its trigger event and recipient.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →