Primary-source verification of recent AI laws, standards and rulings

Verdict key: CONFIRMED / PARTIALLY CONFIRMED / NOT VERIFIABLE / CONTRADICTED.

1. California AB 316: no “AI acted autonomously” defense, in force 1 January 2026

Claim checked: AB 316 is in force from 1 January 2026; a defendant may not avoid liability by claiming the AI acted autonomously.

Primary URL: https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260AB316 (10/13/25, Chaptered; Chapter 672, Statutes of 2025)

Verbatim excerpt (chaptered text):

"Assembly Bill No. 316 CHAPTER 672 An act to add Section 1714.46 to the Civil Code, relating to civil actions. [Approved by Governor October 13, 2025. Filed with Secretary of State October 13, 2025.] … This bill would prohibit a defendant who developed, modified, or used artificial intelligence, as defined, from asserting a defense that the artificial intelligence autonomously caused the harm to the plaintiff."

Verdict: CONFIRMED. The statute adds Civil Code § 1714.46, doing exactly what the claim states. The official Civil Code text records the section as "(Added by Stats. 2025, Ch. 672, Sec. 1. (AB 316) Effective January 1, 2026.)": https://leginfo.legislature.ca.gov/faces/codes_displaySection.xhtml?lawCode=CIV&sectionNum=1714.46

2. Four U.S. jurisdictions activating AI duties on 1 January 2027

2(a) New York: RAISE Act (S6953-B / A6453-B)

Claim checked: the NY RAISE Act activates duties on 1 January 2027.

Primary URLs: - https://www.nysenate.gov/legislation/bills/2025/S6953 (signed into law 19 December 2025; Governor's announcement: https://www.governor.ny.gov/news/governor-hochul-signs-nation-leading-legislation-require-ai-frameworks-ai-frontier-models; Chapter 699, Laws of 2025) - Chapter amendment: https://www.nysenate.gov/legislation/bills/2025/S8828 (S8828/A9449; signed 27 March 2026; Chapter 96, Laws of 2026)

Verbatim excerpt (S8828, amending § 3 of Ch. 699 of 2025; strike-out/underline as enrolled):

"§ 3. This act shall take effect ~~on the ninetieth day after it shall have become a law~~ January 1, 2027."

Verdict: CONFIRMED. The RAISE Act as enacted (Ch. 699 of 2025) and amended (Ch. 96 of 2026, which repealed and replaced General Business Law Art. 44-B, §§ 1420 to 1429) makes the frontier-AI framework publication and 72-hour critical-safety-incident reporting duties operative on 1 January 2027.

Notes: The effective-date clause above was confirmed against the official bill text on nysenate.gov (S8828) and on nyassembly.gov (A9449: https://nyassembly.gov/leg/?default_fld=&leg_video=&bn=A09449&term=2025&Summary=Y&Actions=Y&Text=Y). The duties are publication and reporting duties: the RAISE Act does not contain an audit mandate.

2(b) Illinois: SB 315 (Public Act 104-0538, “Artificial Intelligence Safety Measures Act”)

Claim checked: Illinois SB 315 (mandatory third-party AI audit) activates on 1 January 2027.

Primary URL: https://www.ilga.gov/Legislation/publicacts/view/104-0538 (bill status: https://www.ilga.gov/Legislation/BillStatus?DocNum=315&GAID=18&DocTypeID=SB&SessionID=114&GA=104; approved 6 July 2026)

Verbatim excerpts (enrolled text, P.A. 104-0538):

"Section 99. Effective date. This Act takes effect January 1, 2027."

"(d) Beginning on January 1, 2028 or 90 days after a developer first qualifies as a large frontier developer, whichever is later, a large frontier developer shall annually retain a third party to perform an independent audit of compliance with the requirements of this Section. The third party shall conduct audits consistent with generally accepted auditing standards and best practices…"

Verdict: PARTIALLY CONFIRMED. The audit detail in the claim is wrong by one year. The Act takes effect on 1 January 2027, when the transparency-report and disclosure duties for large frontier developers begin. The mandatory third-party audit itself begins on 1 January 2028, or 90 days after a developer first qualifies, whichever is later.

Notes: The enrolled text above was confirmed against the official text of P.A. 104-0538 on ilga.gov.

2(c) Utah: HB 276, Digital Content Provenance Standards Act (2026 General Session)

Claim checked: the Utah Digital Content Provenance Standards Act activates on 1 January 2027.

Primary URL: https://le.utah.gov/~2026/bills/static/HB0276.html (signed by the Governor 24 March 2026; enrolled text: https://le.utah.gov/Session/2026/bills/enrolled/HB0276.pdf)

Verbatim excerpt (enrolled text):

"Section 15. Section 13-72c-101 is enacted to read: CHAPTER 72c. Digital Content Provenance Standards Act … Section 21. Effective Date. This bill takes effect on January 1, 2027."

Verdict: CONFIRMED. The Act's duties (content-provenance disclosures by generative AI providers and platforms) take effect on 1 January 2027.

Notes: The verbatim text above was confirmed against the enrolled bill on le.utah.gov.

2(d) Colorado: SB 26-189 (repeals and re-enacts the SB 24-205 regime; ADMT disclosure effective 1 January 2027)

Claim checked: Colorado SB 26-189 repealed SB 24-205 before it took effect and created an ADMT disclosure regime effective 1 January 2027.

Primary URLs: - Bill page: https://leg.colorado.gov/bills/sb26-189 - Session law (Chapter 131): https://leg.colorado.gov/laws/session-laws/SB26-189/131/download

Verbatim excerpts:

Session law: "SECTION 1. In Colorado Revised Statutes, repeal and reenact, with amendments, part 17 of article 1 of title 6 as follows: … (1) Except as otherwise provided in subsection (2) of this section, this act takes effect January 1, 2027. … (3) This act applies to consequential decisions made on or after January 1, 2027. … Approved: May 14, 2026"

Bill page summary: "In 2024, the general assembly enacted Senate Bill 24-205, which created consumer protections in interactions with artificial intelligence systems. The act repeals and reenacts those provisions with new requirements regarding the use of automated decision-making technology in consequential decisions."

Verdict: PARTIALLY CONFIRMED. SB 26-189 (Ch. 131, approved 14 May 2026) repeals and reenacts, with amendments, the SB 24-205 provisions (part 17 of article 1 of title 6, C.R.S.). It takes effect on 1 January 2027 and applies to consequential decisions made on or after that date. SB 24-205's own effective date had been delayed to 30 June 2026 by SB25B-004 (https://leg.colorado.gov/bills/sb25b-004). Because the session law makes the repeal and re-enactment take effect on 1 January 2027, the statement that SB 24-205 was repealed before it took effect is not supported by the session law text.

3. EU Digital Omnibus on AI: Regulation (EU) 2026/1744 (the adopted act, not the proposal)

Claim checked: the Digital Omnibus is now law as Regulation (EU) 2026/1744, published in the OJ on 24 July 2026 and in force from 27 July 2026. It defers Annex III high-risk obligations to a 2 December 2027 backstop, and Annex I embedded systems to 2 August 2028.

Primary URLs: - OJ text (EU Publications Office / EUR-Lex cellar, official): https://publications.europa.eu/resource/celex/32026R1744 (canonical ELI: http://data.europa.eu/eli/reg/2026/1744/oj; EUR-Lex page: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32026R1744) - Cellar metadata (dates): same resource, RDF notice.

Verbatim excerpts:

"REGULATION (EU) 2026/1744 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 8 July 2026 amending Regulations (EU) 2024/1689, (EU) 2018/1139 and (EU) 2023/1230 as regards the simplification of the implementation of harmonised rules on artificial intelligence (Digital Omnibus on AI)"

Amended AI Act Article 113: "(c) Chapter III, Sections 1, 2, and 3, with the exception of Article 6(5), shall apply from: (i) 2 December 2027 as regards AI systems classified as high-risk pursuant to Article 6(2) and Annex III; and (ii) 2 August 2028 as regards AI systems classified as high-risk pursuant to Article 6(1) and Annex I;"

"Article 4: Entry into force and application. This Regulation shall enter into force on the third day following that of its publication in the Official Journal of the European Union. … Done at Strasbourg, 8 July 2026."

Cellar/CDM metadata: date_publication = 2026-07-24 (OJ L, 2026/1744); start_of_validity = 2026-07-27; date of signature = 2026-07-08.

Verdict: CONFIRMED. The adopted regulation's OJ page exists, and every claimed date was verified verbatim: - OJ publication: 24 July 2026. - In force: 27 July 2026. - Dates of application for Chapter III, Sections 1 to 3 (fixed dates in the adopted text): 2 December 2027 for Annex III, and 2 August 2028 for Annex I.

The earlier proposal is CELEX 52025PC0836; the adopted act is CELEX 32026R1744.

Notes: The official OJ content (formex XML) and the official metadata were retrieved from publications.europa.eu, the EU Publications Office cellar, which holds the same primary content that EUR-Lex serves. Citation: Regulation (EU) 2026/1744, OJ L, 2026/1744, 24.7.2026, ELI: http://data.europa.eu/eli/reg/2026/1744/oj.

3(b) EU AI Act Article 12(1), Regulation (EU) 2024/1689

Primary URL: http://data.europa.eu/eli/reg/2024/1689/oj (EUR-Lex CELEX 32024R1689; text retrieved from the official OJ formex XML at the publications.europa.eu cellar, OJ L, 2024/1689, 12.7.2024)

Verbatim excerpt (Article 12(1)):

"High-risk AI systems shall technically allow for the automatic recording of events (logs) over the lifetime of the system."

Verdict: CONFIRMED (exact wording as quoted above).

4. NIST CAISI AI Agent Standards Initiative, announced 17 February 2026

Claim checked: NIST CAISI announced its AI Agent Standards Initiative on 17 February 2026.

Primary URL: https://www.nist.gov/news-events/news/2026/02/announcing-ai-agent-standards-initiative-interoperable-and-secure

Verbatim excerpt (nist.gov news page, dated February 17, 2026):

"February 17, 2026: Today, the Center for AI Standards and Innovation (CAISI) at NIST announced the launch of the AI Agent Standards Initiative. The Initiative will ensure that the next generation of AI, AI agents capable of autonomous actions, is widely adopted with confidence, can function securely on behalf of its users, and can interoperate smoothly across the digital ecosystem. Working in coordination with other federal partners, including the Information Technology Laboratory (ITL) at NIST, CAISI aims to foster the emerging ecosystem of industry-led AI standards and protocols…"

Verdict: CONFIRMED.

5. ISO/IEC 24970 (AI system logging)

Claim checked: ISO/IEC 24970 is at DIS stage, its ballot closed in February 2026, and it defines “tamper-evident log”.

Primary URL: https://www.iso.org/standard/88723.html

Verbatim excerpt (iso.org, retrieved 1 October 2026):

"ISO/IEC FDIS 24970, Artificial intelligence, AI system logging … Status: Under development, Stage: Proof sent to secretariat or FDIS ballot initiated: 8 weeks [50.20] … Edition: 1 … Number of pages: 26 … Technical Committee: ISO/IEC JTC 1/SC 42 … Abstract: This document describes common capabilities, requirements and a supporting information model for logging of events in AI systems. This document is designed to be used with a risk management system."

Verdict: PARTIALLY CONFIRMED. The stage detail in the claim is outdated, and “tamper-evident log” cannot be verified from a free primary source. - The standard exists with the claimed title, but as of 1 October 2026 it is at FDIS stage 50.20 (FDIS ballot initiated), not DIS. - iso.org's public page does not publish stage history, so it cannot confirm whether the DIS ballot closed in February 2026. - The draft text is not freely available, so any definition of “tamper-evident log” could not be checked. No primary source for the term was found in this pass.

Notes: Citation: ISO/IEC FDIS 24970, Artificial intelligence, AI system logging (stage 50.20, approval phase), ISO/IEC JTC 1/SC 42. Whether it defines “tamper-evident log” can be confirmed only from the draft or published text.

6. Mendones v. Cushman & Wakefield, Inc. (Alameda County Superior Court, 9 September 2025)

Claim checked: the first terminating sanctions for deepfake evidence, Alameda County Superior Court, 9 September 2025.

Court document: Order re: Terminating Sanctions, Mendones et al. v. Cushman & Wakefield, Inc. et al., No. 23CV028772 (Cal. Super. Ct., Alameda County), Judge Victoria Kolakowski, dated and file-stamped 9/9/2025; the clerk's certificate of electronic service is file-stamped 9/10/2025. - Copy of the signed, file-stamped order, uploaded to DocumentCloud by The National Law Journal: https://www.documentcloud.org/documents/26157672-mendones-v-cushman-wakefield - A court-hosted copy was not found in this pass. - Corroborating legal press and trackers: - Ropes & Gray AI Court Order Tracker (search “Mendones”): https://www.ropesgray.com/en/sites/artificial-intelligence-court-order-tracker?query=Mendones - eDiscovery Today, 25 September 2025: https://ediscoverytoday.com/2025/09/25/deepfake-videos-and-images-lead-to-terminating-sanctions-ediscovery-case-law/

Verbatim excerpt (court order):

"The Court finds that a terminating sanction is appropriate. This sanction is proportional to the harm that Plaintiffs' misuse of the Court's processes has caused. … Further, a terminating sanction serves the appropriate deterrent effect of showing the public that the Court has zero tolerance with attempting to pass deepfakes as evidence. … IV. ORDERS. The Court strikes the second amended complaint from the register of actions. The entire action is dismissed with prejudice. Dated: 9/9/2025, Victoria Kolakowski, Judge of the Superior Court. … [certificate of electronic service:] CASE NUMBER: 23CV028772 … FILED Superior Court of California County of Alameda 09/10/2025."

Ropes & Gray tracker: "In Mendones, et al. vs. Cushman and Wakefield, Inc., et al., No.: 23CV028772 (Cal. Super. Ct. Sept. 9, 2025), the court issued a terminating sanction (i.e., striking the second amended complaint and dismissing the action with prejudice) against plaintiffs for intentionally submitting false evidence involving GenAI deepfake video evidence…"

Verdict: PARTIALLY CONFIRMED. The signed order itself (a copy) and several reputable legal-press sources confirm the case, court, date, judge and terminating-sanctions outcome. That these were the first terminating sanctions for deepfake evidence is not shown by any primary source found in this pass.

7. Proposed FRE 707 and 901(c): not advanced on 7 May 2026

Claim checked: FRE 707 and 901(c), the proposed federal evidence rules on AI, were NOT advanced on 7 May 2026.

Primary URLs: - May 7, 2026 Advisory Committee agenda book (uscourts.gov): https://www.uscourts.gov/sites/default/files/document/2026-05_evidence_rules_agenda_book.pdf (Tab III "Machine Learning and Proposed Rule 707"; Tab IV "Deepfakes and the Federal Rules on Authenticity") - Report of the Advisory Committee on Evidence Rules, May 17, 2026, in the Standing Committee agenda book (June 3-4, 2026), TAB 25: https://www.uscourts.gov/sites/default/files/document/2026-06_standing_committee_meeting_agenda_book_final.pdf - Committee reports page: https://www.uscourts.gov/forms-rules/records-rules-committees/committee-reports/advisory-committee-evidence-rules-may-2026

Verbatim excerpt (Report of the Advisory Committee on Evidence Rules, 17 May 2026, describing the 7 May 2026 meeting):

"The Advisory Committee on Evidence Rules (the 'Committee') met on May 7, 2026, in Washington, D.C. … The Committee does not recommend action on the proposed Rule 707 at this time. Instead, it has revised the proposed Rule and plans to conduct further study on it and another issue relating to artificial intelligence, the problems posed by 'deepfakes', at its next meeting."

On draft Rule 901(c): "Mindful that technology develops quickly and the rule-making process is slow, the Committee's objective is to fine tune a possible amendment to hold in abeyance until such time that it concludes an amendment is warranted…" (901(c) remains an unapproved working draft; it was not published for comment.)

Verdict: CONFIRMED. Neither proposed Rule 707 nor draft Rule 901(c) was advanced at the 7 May 2026 meeting. At the Standing Committee's meeting of 3 and 4 June 2026, the only Evidence Rules item advanced for final approval was the unrelated Rule 609 amendment; Rules 707 and 901(c) appeared as information items only.

Notes: Any claim that an AI evidence rule “takes effect December 1, 2027” is now wrong. That date was only the earliest possible date on the original schedule.

8. California SB 53 (TFAIA), in force 1 January 2026

Claim checked: SB 53, on frontier AI transparency and incident reporting, is in force from 1 January 2026.

Primary URL: https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53 (09/29/25, Chaptered; Chapter 138, Statutes of 2025)

Verbatim excerpt (chaptered text):

"Senate Bill No. 53 CHAPTER 138 An act to add Chapter 25.1 (commencing with Section 22757.10) to Division 8 of the Business and Professions Code, to add Section 11546.8 to the Government Code, and to add Chapter 5.1 (commencing with Section 1107) to Part 3 of Division 2 of the Labor Code, relating to artificial intelligence. [Approved by Governor September 29, 2025. Filed with Secretary of State September 29, 2025.] … This bill would enact the Transparency in Frontier Artificial Intelligence Act (TFAIA) that would … require a large frontier developer to write, implement, and clearly and conspicuously publish on its internet website a frontier AI framework … The TFAIA would require the Office of Emergency Services to establish a mechanism to be used by a frontier developer or a member of the public to report, as prescribed, a critical safety incident, as defined…"

Verdict: CONFIRMED. SB 53 was chaptered on 29 September 2025. The official Business and Professions Code text records § 22757.10 as "(Added by Stats. 2025, Ch. 138, Sec. 2. (SB 53) Effective January 1, 2026.)": https://leginfo.legislature.ca.gov/faces/codes_displaySection.xhtml?lawCode=BPC&sectionNum=22757.10. The digest text confirms the frontier-AI framework publication and critical-safety-incident reporting duties.

9. Texas TRAIGA (HB 149): Attorney General civil investigative demands for AI documentation

Claim checked: the Texas Responsible Artificial Intelligence Governance Act gives the Attorney General civil investigative demands for AI documentation.

Primary URLs: - History: https://capitol.texas.gov/BillLookup/History.aspx?LegSess=89R&Bill=HB149 - Enrolled text: https://capitol.texas.gov/tlodocs/89R/billtext/html/HB00149F.htm

Verbatim excerpts:

History page: "Relating to regulation of the use of artificial intelligence systems in this state; providing civil penalties. … 06/22/2025 E Signed by the Governor … 06/22/2025 E Effective on 1/1/26."

Enrolled text, Sec. 552.103 (Bus. & Com. Code): "INVESTIGATIVE AUTHORITY. (a) If the attorney general receives a complaint through the online mechanism under Section 552.102 alleging a violation of this chapter, the attorney general may issue a civil investigative demand to determine if a violation has occurred. … (b) The attorney general may request from the person reported through the online mechanism, pursuant to a civil investigative demand issued under Subsection (a): (1) a high-level description of the purpose, intended use, deployment context, and associated benefits of the artificial intelligence system …; (2) a description of the type of data used to program or train the artificial intelligence system; … (8) any other relevant documentation reasonably necessary for the attorney general to conduct an investigation under this section."

Also Sec. 552.101(a): "The attorney general has exclusive authority to enforce this chapter…"

Sec. 552.105 sets civil penalties at three levels: - $10,000 to $12,000 for a curable violation; - $80,000 to $200,000 for an uncurable violation; - $2,000 to $40,000 per day for a continuing violation.

Verdict: CONFIRMED. TRAIGA was signed on 22 June 2025 and took effect on 1 January 2026. The Attorney General's power to issue civil investigative demands for AI documentation was verified verbatim.

Summary verdict list