We are not attorneys, and nothing on this site is legal advice. This directory may be incomplete or out of date. Always rely on the official sources and your own legal counsel.
AI Compliance by Design: Official Sources and Book Notes
By R. R. Johnson

Example compliance calendar

This is an example, not a legal schedule. It shows how one fictional organization might spread its recurring AI compliance work across a year. Your own calendar depends on the laws that apply to you and what your counsel advises. Check every date with the official source.

Duties that run every month

ActivityWhenWho leadsEvidence
AI Governance Board regular meeting (special sessions within 3 business days of any pause or escalated incident)MonthlyAI Accountability Leader (chair)07_governance_records: minutes, decision log
Reconcile the six inventory feeds and classify new entries, suspected High uses firstWeekly, with a monthly summary to the boardInventory team01_inventory
Outcome and error monitoring for High usesMonthly or quarterly, as set in each system's monitoring planBusiness owner, with data science06_monitoring_and_incidents
Review open exceptions, open pauses, and overdue conditionsMonthlyAI Accountability Leader07_governance_records: exceptions register

Q1: January to March

ActivityWhenWho leadsEvidence
Confirm compliance with laws that took effect January 1 (many state laws start on this date)First two weeks of JanuaryCounsel, with the AI Accountability Leader07_governance_records: legal change log
Quarterly business owner attestation that every use entry is currentJanuaryBusiness owners01_inventory: signed attestations
Quarterly review of every High use and every agent with write accessJanuary to FebruaryAI Accountability Leader03_systems; 05_vendors_and_security
Quarterly AI report to the board committee (the five map numbers, log readiness, open pauses)FebruaryAI Accountability Leader07_governance_records: board papers
Annual review of the AI policy, the acceptable use policy, and the classification rubric against new lawFebruary to MarchAI Accountability Leader, with counsel02_policies: version history
Annual AI literacy and acceptable use training refresh for all staff, and role-based training for reviewers and agent ownersLaunch in Q1, complete by the end of Q2HR and learning, with the AI Accountability Leader07_governance_records: training records
Review AI claims and AI risk disclosures in the annual report, marketing, and investor materials (public companies and regulated firms)Before the annual report is filedCounsel and the disclosure committee02_policies: claims register
Prepare California risk assessment summaries for the April 1 submission (from 2028)February to MarchPrivacy office, with counsel03_systems: risk assessments

Q2: April to June

ActivityWhenWho leadsEvidence
California submission of risk assessment information to the CPPA. The first submission, for assessments done in 2026 and 2027, is due April 1, 2028, and then each April 1April 1Privacy office03_systems: submission receipt
California cybersecurity audit certification, phased by revenue: April 1, 2028 (over $100 million), April 1, 2029 ($50 to $100 million), April 1, 2030 (under $50 million)April 1 of the applicable yearChief Information Security Officer05_vendors_and_security
Quarterly business owner attestationAprilBusiness owners01_inventory
Quarterly AI report to the board committeeMayAI Accountability Leader07_governance_records
Annual bias testing for High hiring and promotion tools, and the NYC Local Law 144 independent bias audit (must be no more than one year old at the time of use, with a public summary). Schedule it ahead of each tool's anniversary, not by quarterAt least yearly; Q2 in this exampleHR, with an independent auditor03_systems: test results; published summary
Track state legislative sessions as they close and log new AI bills signed into law (many legislatures adjourn in the spring)April to JuneCounsel07_governance_records: legal change log
Mid-year incident response tabletop exercise, including an Agentic AI scenarioJuneAI Accountability Leader, with security06_monitoring_and_incidents

Q3: July to September

ActivityWhenWho leadsEvidence
Quarterly business owner attestationJulyBusiness owners01_inventory
EU AI Act date check. August 2 is a recurring EU milestone: transparency duties applied from August 2, 2026, and Annex I product rules apply from August 2, 2028July, ahead of August 2Counsel07_governance_records: legal change log
Annual vendor AI feature questionnaire: ask every vendor of an inventoried system to list the AI features added in the past yearJuly to AugustProcurement, with the AI Accountability Leader05_vendors_and_security
Quarterly AI report to the board committeeAugustAI Accountability Leader07_governance_records
Annual review of every Minimal and Limited useAugust to SeptemberInventory team, with business owners01_inventory
Annual internal audit of the AI governance program, using the Audit Guide (and, for certified organizations, the ISO/IEC 42001 surveillance audit)September, or as your audit plan setsInternal audit07_governance_records: audit reports

Q4: October to December

ActivityWhenWho leadsEvidence
Quarterly business owner attestationOctoberBusiness owners01_inventory
Targeted re-classification before the year-end and January 1 legal datesOctober to NovemberInventory team, with counsel01_inventory: classification records
EU December 2 deadlines: new prohibitions and the deadline for legacy synthetic-content marking (December 2, 2026); Annex III high-risk rules, including fundamental rights impact assessments where required (December 2, 2027)Before December 2 of the applicable yearCounsel and system owners03_systems; 07_governance_records
California risk assessments for ADMT and other processing already under way by December 31, 2027; after that, review each assessment at least every three years and within 45 days of a material changeDecember 31, 2027, then ongoingPrivacy office03_systems: risk assessments
Quarterly AI report to the board committee, plus the annual program report and next year's plan and budgetNovemberAI Accountability Leader07_governance_records
Readiness check for laws starting January 1, including Colorado SB 26-189 and the Colorado Chatbot Safety Act (HB 26-1263) and California's ADMT rules, all from January 1, 2027November to DecemberCounsel02_policies; 07_governance_records
Annual records retention review: confirm logs and decision records are kept for the longest period any applicable law requiresDecemberRecords management, with counsel06_monitoring_and_incidents

Dated deadlines, as printed in the book, current as of October 19, 2026

This page does not keep these dates current. Follow the source link for each one.

DateWhat happensWho it affectsOfficial source
Already in forceTexas TRAIGA prohibitions and disclosure duties (January 1, 2026); Illinois HB 3773 employment rules (January 1, 2026); NYC Local Law 144 (enforced since July 5, 2023); EU prohibitions (February 2, 2025), general-purpose AI duties (August 2, 2025), and Article 50 transparency duties (August 2, 2026)Anyone using AI in those placesTX-001, IL-001, NYC-001, EU-001
October 26, 2026Colorado Attorney General written comment deadline and formal hearing on the ADMT rulesOrganizations making consequential decisions about Colorado consumersCO-001
December 2, 2026EU: new prohibitions on non-consensual intimate imagery and child sexual abuse material; legacy generative systems must meet the Article 50(2) marking ruleProviders and deployers in the EU marketEU-001, EU-002
January 1, 2027Colorado SB 26-189 and HB 26-1263 take effect (enforcement of SB 26-189 is under a court-approved stay; check its status); California ADMT duties for existing uses beginDeployers and developers serving Colorado and California consumersCO-001, CA-001
December 2, 2027EU Annex III high-risk rules applyProviders and deployers of high-risk systems in the EUEU-001
December 31, 2027California: first risk assessments due for processing already under wayBusinesses subject to the CCPACA-001
April 1, 2028California: first risk assessment submissions and the first cybersecurity audit certifications (largest businesses)Businesses subject to the CCPACA-001
August 2, 2028EU Annex I high-risk rules for AI in regulated products applyProduct manufacturers and their AI suppliersEU-001
April 1, 2029 and April 1, 2030California cybersecurity audit certifications for the next two revenue tiersBusinesses subject to the CCPACA-001

Adapted from Appendix F, Part 2 of AI Compliance by Design.