Regulation Tracker · 20 questions
How courts have applied costs, contempt, referral and disciplinary powers to AI misuse, where no written rule exists.
As of 2026-09-04, 1 of 130 countries and entities have a binding rule for sanctions for AI misuse, 1 have official guidance, 0 have a proposal, 32 are set by case law, 88 were checked with nothing found and 8 remain unclear. Binding countries: Ireland.
Tap or hover a country. Faint land has no record yet. Miller projection; boundaries as published by Natural Earth (India point of view) — see the disclaimer.
Europe · verified 2026-09-04
PD HC 142 sets out the sanctions expressly: rejection of documents, adverse or personal costs orders, strike-out and referral to the regulator. Before it, the Court of Appeal dismissed both Guerin v O'Doherty [2026] IECA 48 and Von Geitz v Kelly [2026] IECA 29 on submissions containing non-existent authorities.
International bodies · verified 2026-09-04
No international tribunal has sanctioned AI misuse. Ciarb 7.6-7.7 lets tribunals draw adverse inferences and reflect undisclosed AI use in costs. The only reported challenge is domestic to the United States: LaPaglia v Valve, filed 8 April 2025, alleging an arbitrator outsourced the award to AI; the outcome is unverified.
No countries in this status.
Americas · verified 2026-09-04
The Rosario appeals court issued a warning and referred the lawyer to the Rosario bar in August 2025; a General Roca appeals court found 26 fabricated citations; a Neuquen court referred a lawyer to the Cutral Co bar. The Rio Negro acordada of 1 October 2025 authorises corrective sanctions and mandatory referral to bar ethics tribunals.
Asia-Pacific · verified 2026-09-04
The largest body of AI-citation case law outside the United States (110 Australian decisions in the Charlotin database at 4 September 2026). Courts use personal costs orders and referral to regulators: Valu (No 2), Dayal (VLSB+C practising-certificate conditions), JNE24 (AUD 8,371 personal costs), Asif (AUD 9,098 against a litigant).
Americas · verified 2026-09-04
A consistent line since 2025 under articles 80 and 81 of the civil procedure code: a Santa Catarina warning in February 2025; a Londrina federal court fine of 20 minimum wages plus bar referral in July 2025; a labour court bad-faith fine in March 2026; and a Parana 2 per cent fine with bar referral in June 2026.
Americas · verified 2026-09-04
Courts apply existing costs and contempt powers. Zhang v. Chen, 2024 BCSC 285 awarded costs against counsel who filed fabricated ChatGPT authorities; Ko v. Li (Ont. S.C.J. 2025) opened contempt proceedings; the Federal Court and Ontario courts have made similar orders. No AI-specific sanctions rule exists.
Americas · verified 2026-09-04
Courts use their disciplinary power under article 531 of the Organic Code of Courts, and article 542 for the Supreme Court: Rol 23.322-2025 suspended a lawyer for one month and fined 5 UTM on 22 April 2026, and the second civil court of Concepcion fined 1 UTM in February 2026.
Asia-Pacific · verified 2026-09-04
The Tongzhou District court criticised counsel in the judgment without a fine because the consequences were minor; Case Database guidance from December 2025 anticipates fines under Civil Procedure Law article 114 for serious cases, and Dawu County issued an admonishment for AI-fabricated evidence in November 2025.
Americas · verified 2026-09-04
The Supreme Court annulled a Sincelejo tribunal ruling built on non-existent citations as a due-process violation on 5 November 2025 and fined a lawyer 15 minimum monthly wages; the Council of State referred a lawyer to the Bogota sectional commission of judicial discipline in June 2026.
Europe · verified 2026-09-04
Constitutional Court, I. ÚS 3004/25 (1 Dec 2025), fined counsel Mgr. Pavel Kehl CZK 25,000 — a quarter of the CZK 100,000 maximum, as a first offence — over a constitutional complaint citing non-existent decisions (III. ÚS 2987/18, III. ÚS 3881/18, II. ÚS 2264/14). Anglicisms and inconsistent citation formatting indicated AI drafting.
Europe · verified 2026-09-04
Advokatnævnet fined a lawyer DKK 10,000 payable to the Treasury in December 2025 (reported 19 May 2026) for citing two non-existent judgments in an email; the errors spanned year, page, court and content. The ruling makes no finding about AI and the lawyer denied using it, but it was logged as Denmark's first Hallucination Cases entry.
Europe · verified 2026-09-04
At least eight French decisions since December 2025 have flagged generative-AI misuse. TA Orléans (29 Dec 2025, no. 2506461) censured a represented applicant's counsel for about fifteen wholly fabricated references; TJ Périgueux (pôle social, 18 Dec 2025) found fictitious case references in submissions. So far the courts have issued judicial warnings and rejected the submissions; no bar disciplinary sanction and no article 32-1 CPC amende civile has been reported.
Europe · verified 2026-09-04
Four reported decisions, all short of a formal sanction. AG Köln 2 Jul 2025 (312 F 130/25): every source from page 8 onwards invented; conduct impedes fact-finding and damages the standing of the profession (§ 43a(3) BRAO). OLG Celle 29 Apr 2025 (5 U 1/25): pleading built on invented OLG citations. LG Frankfurt 25 Sep 2025 (2-13 S 56/24): fabricated BGH quotation disregarded and the Streitwert set at EUR 3,000. KG Berlin 20 Nov 2025 (17 WF 144/25): formal admonition. No costs order or disciplinary measure reported.
Asia-Pacific · verified 2026-09-04
Two reported decisions, both against unrepresented parties and both ending without sanction: Yu Hon Tong Thomas v Centaline Property Agency (High Court, 26 February 2025) and Licksun Company Limited v Occupiers of Lot No. 552 (District Court, 8 August 2025), where six lodged authorities were found not to exist and the court held the lodging had the effect of misleading it.
Asia-Pacific · verified 2026-09-04
Pooja Ramesh Singh (2 July 2026) set aside orders resting on six hallucinated precedents and declared unverified AI citation advocate misconduct; Gadiya (2 September 2026) set aside a Rs 425.27 crore customs penalty and recommended disciplinary action. Earlier decisions came from the Delhi and Bombay High Courts and ITAT Bengaluru.
Middle East · verified 2026-09-04
Courts use costs and professional censure. Supreme Court, Justice Gila Canfy-Steinitz, Feb 2025: petition dismissed where counsel cited 36 non-existent or misquoted judgments. Followed by the Ramat Gan, Adibi and Haifa Labour Court sanctions. No dedicated sanctions rule exists.
Europe · verified 2026-09-04
Tribunale di Siracusa judgment 338/2026 applied article 96(3)-(4) of the code of civil procedure for four phantom Cassazione citations, awarding about 30,200 euro in fees, damages and a court-fund payment. Tribunale di Firenze refused sanctions absent bad faith and TAR Lombardia referred counsel to the bar.
Africa · verified 2026-09-04
In the 16 Apr 2026 ruling the High Court set aside the 23 Dec 2025 judgment and struck out the AI-founded application with costs against the applicant. The March 2026 ruling struck out the filing without costs sanction. No disciplinary proceedings against an advocate for AI misuse were found.
Europe · verified 2026-09-04
Raad van Discipline 's-Hertogenbosch, 27 Jul 2026, ECLI:NL:TADRSHE:2026:93 — the first Dutch AI discipline case. A lawyer filed submissions in two rental disputes with fabricated or mismatched ECLI numbers (one pointing to a divorce ruling, another to a criminal case), then filed rectification acts without replacement authority. Breach of deskundigheid and integriteit (art. 10a Advocatenwet) and Gedragsregel 8. Sanction: berisping plus EUR 1,250 costs (EUR 750 NOvA, EUR 500 State); the confidentiality charge failed because no client data had been entered.
Asia-Pacific · verified 2026-09-04
Seven reported New Zealand decisions, all involving self-represented litigants, with warnings rather than penalties. In Jones v Family Court at Whangarei [2026] NZSC 1 the Supreme Court awarded NZ$500 costs and warned that reliance on unverified AI output may in serious cases amount to obstruction of justice or contempt.
Europe · verified 2026-09-04
In April 2025 the Supreme Court received its first brief with AI-fabricated statutes and preparatory works. Chief Justice Toril Marie Øie confirmed the court contacted the lawyer without sanctioning, but said future breaches could bring procedural fines [rettergangsbot] or referral to the disciplinary board.
Asia-Pacific · verified 2026-09-04
Thin and unpublished. In a Sandiganbayan Seventh Division minute resolution promulgated 7 August 2025 (People v Cunanan), counsel was reminded to fact-check AI-generated research and observe proper citations under the CPRA after non-existent authorities and wrong docket numbers appeared in AI-drafted motions. Warning only; no official URL, as minute resolutions are unpublished.
Europe · verified 2026-09-04
The District Court in Wrocław dismissed a ChatGPT-drafted claim that mixed the Civil Code with public-procurement law and subtly altered quoted provisions. No reported Polish disciplinary sanction against a lawyer for hallucinated citations as of 4 September 2026.
Europe · verified 2026-09-04
The Guimaraes Court of Appeal on 10 February 2026 ordered 5 UC in costs, about 500 euro, and suggested bar referral for six invented precedents. On the bench side, the CSM opened an inquiry in February 2025 into three Lisbon appeal judges over a judgment in which twelve of fourteen cited rulings did not exist.
Middle East · verified 2026-09-04
In Sheppard v Jillion LLC [2025] QIC (E) 3 (12 Nov 2025) the QFC Civil and Commercial Court, per Lord Thomas of Cwmgiedd, found a Dubai-based lawyer in contempt under Article 35 for citing two non-existent QFC authorities and persisting after challenge, imposing no penal sanction but publishing the judgment.
Asia-Pacific · verified 2026-09-04
Seven reported decisions and the region's most consistent use of money sanctions. Tan Hai Peng [2026] SGHC 49 (S$10,000 across two practitioners); Tajudin bin Gulam Rasul (S$800 personal costs plus an order to inform the client, 29 September 2025); Goh Chin Cheng v Choco Up SG (S$900, not to be passed to the client, 4 May 2026); adverse costs of S$30,000 and S$18,947 against litigants.
Europe · verified 2026-09-04
Reported: Supreme Court of the Slovak Republic, 4CdoNa/3/2026 (31 Mar 2026), imposed no sanction but recorded suspicion of AI drafting from incomplete sentences, holding the party bears full responsibility and must answer for any hallucinating AI arguments. Sourced from the SAK Bulletin; not verified against the official decisions database.
Africa · verified 2026-09-04
Courts apply existing costs, contempt and referral powers. Mavundla v MEC (KZN) and Northbound Processing dealt with practitioners' fabricated citations; in FJL v TGO (Gauteng full court, 31 July 2026) Opperman J referred an acting judge to the LPC over at least eleven fictitious citations; Molawa v Matjhabeng (Labour Court, 13 July 2026) concerned a disciplinary chairperson's hallucinated authorities.
Europe · verified 2026-09-04
TSJ Navarra issued a warning without sanction in September 2024. TSJ Canarias fined a lawyer about 420 to 450 euro under article 247 of the civil procedure law for 48 non-existent Supreme Court citations in February 2026. TSJ Navarra archived article 247 proceedings in April 2026 after an apology. Circular 3/2026 adds discipline.
Europe · verified 2026-09-04
Eskilstuna District Court, judgment reported 13 Jan 2026, criticised counsel for 'AI slop' pleadings citing four non-existent Supreme Court judgments and a misdescribed Contracts Act provision, and reduced the claimed fee. Case number not published in available reports.
Middle East · verified 2026-09-04
Sanctions run through existing criminal-complaint and bar disciplinary powers. Kızılcahamam civil court, June 2026: fabricated Court of Cassation case numbers in a pleading triggered a court criminal complaint plus an Ankara Bar investigation of lawyer E.Ç. Istanbul, Aug 2026: a non-existent Constitutional Court decision cited in a TRY 10m promissory-note case.
Middle East · verified 2026-09-04
In Arabyads Holding Ltd v Gulrez Alam Marghoob Alam [2025] ADGMCFI 0032 (18 Dec 2025), Justice Paul Heath KC ordered the defendant's former representatives, MIO Legal Consultants LLP, to pay AED 282,508 in wasted costs on an indemnity basis for a 327-paragraph defence containing fictitious and miscited authorities. The conduct was found reckless and a breach of the ADGM Rules of Conduct.
Europe · verified 2026-09-04
Ayinde v Haringey and Al-Haroun (Divisional Court, June 2025) set the framework: wasted costs, contempt, regulatory referral. Applied since in Bandla v SRA [2025] EWHC 1167 (Admin) (GBP 24,727.20 costs against a former solicitor who cited about 27 non-existent authorities) and Folarin [2026] UKFTT 135 (GRC). The UK remains the most enforcement-active jurisdiction in this group; German commentary (Anwaltsblatt) contrasts it with the largely consequence-free German position.
Americas · verified 2026-09-04
Federal courts sanction AI misuse under Rule 11, 28 U.S.C. 1927 and inherent power. Mata v. Avianca (S.D.N.Y. 2023) set the pattern; Withers v. City of Aberdeen (N.D. Miss., June 2026) removed all four counsel for fabricated authorities; by mid-2026 more than a thousand US decisions had addressed AI-fabricated material.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
Europe · verified 2026-09-04
No reported Austrian court or disciplinary decision on hallucinated AI citations found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
Middle East · verified 2026-09-04
No rule, guidance or reported case found as of 4 September 2026. No Bahraini decision sanctioning a lawyer for AI-fabricated citations was located.
Asia-Pacific · verified 2026-09-04
No reported case of AI-fabricated citations before Bangladeshi courts found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
Europe · verified 2026-09-04
No reported Belgian court or disciplinary decision on hallucinated AI citations found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Americas · verified 2026-09-04
No reported sanction for AI-hallucinated content found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Europe · verified 2026-09-04
No Croatian decision on hallucinated citations found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
Europe · verified 2026-09-04
No Cypriot decision on hallucinated citations found as of 4 September 2026.
Americas · verified 2026-09-04
No reported sanction for AI-hallucinated content found as of 4 September 2026.
Americas · verified 2026-09-04
No reported sanction for AI-hallucinated content found as of 4 September 2026; the July 2026 bill would create a disciplinary offence.
Africa · verified 2026-09-04
No rule, guidance or reported case found as of 4 September 2026. No Egyptian decision sanctioning a lawyer for fabricated AI citations was located in Arabic or English sources.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Europe · verified 2026-09-04
No EU-level sanctions framework for AI misuse in litigation; sanctions for hallucinated citations are imposed by national courts and bars. AI Act penalties (Arts. 99–101) target providers/deployers of AI systems, not litigation conduct.
No rule, guidance or reported case found as of 4 September 2026.
Africa · verified 2026-09-04
no rule, guidance or reported case found as of 4 September 2026
Europe · verified 2026-09-04
No Greek court decision on hallucinated citations found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
Europe · verified 2026-09-04
No Icelandic court sanction, costs order or bar disciplinary decision over hallucinated citations found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No Indonesian court decision or Dewan Kehormatan disposition sanctioning AI-fabricated citations was found as of 4 September 2026; the Charlotin database contains no Indonesian entry, and Indonesian coverage of AI-hallucination sanctions concerns United States cases.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No reported Japanese court sanction for AI-fabricated citations found as of 4 September 2026; Japanese coverage concerns foreign cases.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
Middle East · verified 2026-09-04
No rule, guidance or reported case found as of 4 September 2026. No Kuwaiti decision sanctioning a lawyer for AI-fabricated citations was located.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Europe · verified 2026-09-04
No reported Luxembourg court or disciplinary decision on hallucinated AI citations found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No Malaysian decision sanctioning a lawyer for fabricated AI citations was found as of 4 September 2026; the Charlotin AI Hallucination Cases database (2,009 decisions across 41 jurisdictions, updated 3 September 2026) contains no Malaysian entry. Chief Justice Wan Ahmad Farid has warned publicly against AI over-reliance but issued no directive.
Europe · verified 2026-09-04
No Maltese decision on AI-generated or hallucinated content found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
Africa · verified 2026-09-04
No rule, guidance or reported case found as of 4 September 2026. No Moroccan decision sanctioning a lawyer for AI-fabricated citations was located.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No reported case of AI-fabricated citations in Nepali courts as of 4 September 2026; the Kathmandu Post reported in May 2026 that similar issues have not surfaced in Nepal yet.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
Africa · verified 2026-09-04
No reported Nigerian decision applying costs, contempt or disciplinary powers to AI-fabricated citations was found. No Legal Practitioners Disciplinary Committee ruling on AI located as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No reported costs, contempt or disciplinary order for AI-fabricated citations found as of 4 September 2026. The 2023 Phalia sessions-court use of GPT-4 was a judge's experiment, not a sanction.
Americas · verified 2026-09-04
No reported sanction for AI-hallucinated content found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
Americas · verified 2026-09-04
No reported Peruvian sanction for AI-hallucinated citations found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Middle East · verified 2026-09-04
No reported Saudi court decision sanctioning fabricated AI citations found as of 4 September 2026. Saudi judgments are not systematically published, so absence of reports is weak evidence.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
Europe · verified 2026-09-04
No Slovenian decision on hallucinated citations found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No reported case of AI-fabricated citations before Sri Lankan courts found as of 4 September 2026.
Europe · verified 2026-09-04
No reported Swiss court or disciplinary decision on hallucinated AI citations found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No reported Taiwanese court sanction for AI-fabricated citations found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No Thai decision involving AI-fabricated citations was found as of 4 September 2026. The Charlotin AI Hallucination Cases database (2,009 decisions across 41 jurisdictions, updated 3 September 2026) contains no Thai entry.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No rule, guidance or reported case found as of 4 September 2026.
Americas · verified 2026-09-04
No reported sanction for AI-hallucinated content found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
No reported sanction for AI-hallucinated citations or AI-fabricated legal content found as of 4 September 2026.
Asia-Pacific · verified 2026-09-04
No Vietnamese decision involving AI-fabricated citations was found as of 4 September 2026; the Charlotin AI Hallucination Cases database contains no Vietnamese entry.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
No decision sanctioning a lawyer or litigant for AI-fabricated citations or AI misuse in filings was found as of 4 September 2026.
Europe · verified 2026-09-04
No reported Estonian court sanction or bar disciplinary decision over hallucinated citations was located, but Estonian case databases could not be searched during research.
Europe · verified 2026-09-04
No reported Finnish court sanction or Valvontalautakunta decision over hallucinated citations was located, but case databases could not be searched during research. Position not determined as of 4 September 2026.
Europe · verified 2026-09-04
No Hungarian court sanction or disciplinary decision for hallucinated citations was found, but no case-law search was possible. Absence is not established.
Europe · verified 2026-09-04
No reported Latvian court sanction or Disciplinary Cases Commission decision over hallucinated citations was located, but Latvian case and disciplinary databases could not be searched during research.
Europe · verified 2026-09-04
No reported Lithuanian court sanction or Court of Honour decision over hallucinated citations was located, but Lithuanian case and disciplinary databases could not be searched during research.
Americas · verified 2026-09-04
No reported Mexican decision sanctioning fabricated AI citations was located, but search coverage of Mexican case law was limited in this pass; treat as not established rather than absent.
Europe · verified 2026-09-04
No Romanian court sanction or disciplinary decision for hallucinated citations was found, but no case-law search was possible. Absence is not established.
Asia-Pacific · verified 2026-09-04
The research budget was exhausted before Korean case law on AI-fabricated citations could be checked.
SafeLegalAI Legal AI Regulation Tracker, "Sanctions for AI misuse" by country, https://safelegalai.com/regulation/category/sanctions-record (accessed 2026-09-04). Data: CC BY 4.0.
SafeLegalAI is a research publication by Cognesio LLP, not a law firm. Nothing here is legal advice, and no lawyer–client relationship arises from reading it. Rules change; always check the official document linked on each record and take advice on your own situation. Researched and drafted with AI assistance; verified against primary sources and edited by Cognesio LLP. The linked official documents are the record — our summaries are not the law and are not legal advice.
Ireland
International bodies
Argentina
Australia
Brazil
Canada
Chile
China
Colombia
Czechia
Denmark
France
Germany
Hong Kong SAR
India
Israel
Italy
Kenya
Netherlands
New Zealand
Norway
Philippines
Poland
Portugal
Qatar
Singapore
Slovakia
South Africa
Spain
Sweden
Türkiye
United Arab Emirates
United Kingdom
United States
Afghanistan
Albania
Algeria
Armenia
Austria
Azerbaijan
Bahrain
Bangladesh
Belarus
Belgium
Bolivia
Bosnia and Herzegovina
Botswana
Brunei
Bulgaria
Cambodia
Cameroon
Costa Rica
Côte d'Ivoire
Croatia
Cuba
Cyprus
Dominican Republic
Ecuador
Egypt
El Salvador
Ethiopia
European Union
Georgia
Ghana
Greece
Guatemala
Honduras
Iceland
Indonesia
Iran
Iraq
Jamaica
Japan
Jordan
Kazakhstan
Kuwait
Kyrgyzstan
Laos
Lebanon
Luxembourg
Macao
Malaysia
Malta
Mauritius
Moldova
Mongolia
Montenegro
Morocco
Mozambique
Myanmar
Namibia
Nepal
Nicaragua
Nigeria
North Macedonia
Oman
Pakistan
Panama
Paraguay
Peru
Puerto Rico
Russia
Rwanda
Saudi Arabia
Senegal
Serbia
Slovenia
Sri Lanka
Switzerland
Taiwan
Tanzania
Thailand
Trinidad and Tobago
Tunisia
Uganda
Ukraine
Uruguay
Uzbekistan
Venezuela
Vietnam
Zambia
Zimbabwe
Estonia
Finland
Hungary
Latvia
Lithuania
Mexico
Romania
South Korea