HOME/JURISDICTIONS/FR

FR — Country Profile

France

54TOTAL
15OFFICIAL SOURCES
18TOPIC AREAS
Law / Act7
Policy / Guidance2
National Strategy3
Standard / Framework3
International Agreement4
Working Paper2
Court Case8
News / Press5
Other20
Agentic AiAntitrust & CompetitionChips & Data CentresComputeCopyright & IpCybersecurityData Privacy & ProtectionDeepfakesEnergyEnvironment & ClimateGenerative AIHuman Rights & EthicsJudicial & Law EnforcementLiability & AccountabilityNational StrategyOnline Safety & Child ProtectionSandboxTrade & Investment
Court Case✓ Official

n° 25BX02906 (M. B. c. Préfet de la Vienne)

Lawyer appeared before the CAA Bordeaux. Fabricated: Case Law | Le conseil a cité « CE, 7 février 2018, n° 409302 », que la cour a constaté comme n'existant pas et a invité à vérifier la référence.

Court: CAA BordeauxParty: Lawyer
26 February 2026Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Court Case✓ Official

No. 25BX02906

Lawyer appeared before the CAA Bordeaux. Fabricated: Case Law | Counsel cited 'CE, 7 février 2018, n° 409302'; the court states this decision does not exist.

Court: CAA BordeauxParty: Lawyer
26 February 2026Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Court Case✓ Official

Mme Y

Pro Se Litigant appeared before the TA Rennes.

Court: TA RennesParty: Pro Se Litigant
30 January 2026Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Court Case✓ Official

SAS La Volumerie

Lawyer appeared before the TA Rennes. Fabricated: Case Law

Court: TA RennesParty: Lawyer
28 January 2026Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Court Case✓ Official

N° 2506461

Lawyer appeared before the TA Orléans. Fabricated: Case Law | Court states this cited CAA Lyon decision with number/date does not exist or numbers do not correspond to dates. Outcome: Admonishment.

Court: TA OrléansParty: Lawyer
29 December 2025Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Court Case✓ Official

M. [C] c/ CAF de la [Dpt CAF]

Lawyer appeared before the TJ Périgueux. Fabricated: Case Law | The claimant cited jurisprudence references that did not correspond to published decisions; the court flagged the references as likely AI-generated 'hallucinations' and gave pourvoi n°16-26694 as an example of a misreferenced decision. Outcome: Warning.

Court: TJ PérigueuxParty: Lawyer
18 December 2025Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Court Case✓ Official

Case n° 2512468

Pro Se Litigant used Unidentified in proceedings before the TA Grenoble.

Court: TA GrenobleParty: Pro Se LitigantTool: Unidentified
9 December 2025Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Court Case✓ Official

N° 2509827

Pro Se Litigant appeared before the TA Grenoble. Fabricated: Case Law

Court: TA GrenobleParty: Pro Se Litigant
3 December 2025Judicial & Law EnforcementGenerative AILiability & Accountability
↗ Link available
Other

President Macron addresses India AI summit, saying Europe is determined to 'shape rules' with allies

It is reported that French President Emmanuel Macron addressed the AI Impact Summit in New Delhi, emphasizing Europe's commitment to shaping global AI regulations in collaboration with allies like India. President Macron highlighted the EU's leadership through the AI Act, adopted in 2024, and underscored Europe's dual focus on innovation and safety. President Macron praised India's advancements in digital public infrastructure and advocated for a 'third way' in AI development, independent of US and Chinese dominance. President Macron also stressed the importance of protecting children from digital abuse and announced France's initiative to ban social media use for individuals under 15, aligning with similar measures in Australia.

19 February 2026Human Rights & EthicsOnline Safety & Child Protection
↗ Link available
National Strategy

French President Macron urges Europe to simplify its regulations to get back into the AI race

It is reported that at the Paris AI Action Summit, French President Emmanuel Macron made statements (1) revealing a €109 billion investment package for the French AI ecosystem; (2) emphasizing France's attractiveness for data center projects due to its nuclear energy surplus; (3) framing the summit as a call for a broader European AI strategy, urging the EU to simplify regulations, deepen the single market, and invest in computing capacities; and (4) stressing the need for Europe to synchronize with global standards in various sectors and encouraged European companies to support homegrown AI solutions. The French President's statements come ahead of the expected announcement of the European AI strategy by President Van der Leyen, setting the stage for potential shifts in Europe's approach to AI development and regulation.

10 February 2025National Strategy
↗ Link available
News / Press

France AI summit commences

It is reported that world leaders and tech executives gathered in Paris, France for the AI Action Summit to discuss new investments in AI and how to safely embrace AI amid growing resistance to regulations. The summit highlighted a shift in focus from previous meetings, with increased pressure on the European Union to adopt a lighter-touch approach to AI regulation to maintain competitiveness. French President Emmanuel Macron has called for a balance between regulation and progress. The summit saw the launch of Current AI (a partnership of countries such as France and Germany and industry players including Google and Salesforce) aiming to invest up to $2.5 billion over five years in public-interest AI projects. Additionally, France announced private sector investments totalling 109 billion euros during the event. Top political leaders including U.S. Vice President JD Vance and China's Vice Premier Zhang Guoqing will attend the summit.

10 February 2025Antitrust & CompetitionData Privacy & Protection
↗ Link available
Other

French MPs want to amend EU's copyright rules to cover generative AI

A commission of the French National Assembly published an opinion recommending amending the EU's Copyright Directive so that it takes into account impacts of generative AI on intellectual and industrial properties.

20 January 2024Copyright & IpGenerative AI
↗ Link available
Law / Act

France, Germany, Italy push for 'mandatory self-regulation' for foundation models in EU's AI law

The three biggest EU countries are pushing for codes of conduct without an initial sanction regime for foundation models rather than prescriptive obligations in the AI rulebook.

19 November 2023Generative AISandbox
↗ Link availableSecondary source
Policy / Guidance

French Data Protection Authority publishes guidance on deepfakes including protection and reports of illegal content

The French Data Protection Authority (CNIL) has published guidance on deepfakes explaining the associated privacy and security risks and setting out steps for individuals to protect themselves and report illicit content. The guidance highlights that creating or sharing certain deepfakes can lead to criminal liability under French law, including that creating an image montage of a person without consent may be punishable by one year of imprisonment and a EUR 15,000 fine, and that fraud involving deception to obtain money, property or a service may be punishable by 5 years of imprisonment and a EUR 375,000 fine. The guidance also describes the CNIL’s role in informing the public, supervising compliance with data protection rules, supporting research and detection work, including through the GenFakes project, and contributing to European initiatives, including work on codes of practice for AI-generated content.

3 February 2026Data Privacy & ProtectionDeepfakes
↗ Link available
Other

Competition Authority inquires into competitive state of the conversational agent sector

France's Competition Authority has initiated a public consultation regarding an ex officio inquiry into the competitive functioning of the chatbot sector in France. The inquiry aims to map the current competitive landscape, excluding search engine integrations, to identify major market participants and evaluate significant barriers to entry such as access to massive datasets, high-performance computing power, and substantial capital requirements. Key areas of focus include the risks of vertical integration by tech giants—specifically how embedding chatbots into office suites or messaging platforms might impact interoperability under the Digital Markets Act (DMA) and the AI Act—alongside an exploration of monetization models like token billing and advertising. The Authority is also investigating the emergence of "agentic trade," where AI assistants act as autonomous intermediaries for commercial transactions, raising concerns regarding algorithmic collusion, brand visibility, and the necessity of universal interoperability standards. The collected data and stakeholder recommendations will ultimately inform a formal opinion on the adequacy of the existing legislative framework to govern the rapidly evolving generative AI sector. The consultation closes on 6 March 2026

29 January 2026Antitrust & CompetitionGenerative AIAgentic Ai
↗ Link available
Other

AI chatbots to be scrutinized by French antitrust sector inquiry

France's competition authority (the Autorité de la concurrence) has initiated an ex officio inquiry into the competitive dynamics of the conversational agent sector, focusing on AI-driven tools like ChatGPT, Google Gemini, Mistral AI's Le Chat, Perplexity, and Microsoft's Copilot. This move follows their previous analysis of the generative AI sector and aims to assess the impact of conversational agents on various economic sectors, particularly in the rapidly growing "agentic commerce" within e-commerce. The inquiry will explore issues such as advertising integration, service partnerships, and the evolution of these agents into platforms offering direct access to third-party services. A public consultation will be launched soon, with the final opinion expected in 2026.

9 January 2026Agentic AiAntitrust & CompetitionGenerative AI
↗ Link available
News / Press

France to prosecute X over Grok AI sexual deepfakes

It is reported that French prosecutors have initiated a criminal investigation into the social media platform X following the proliferation of non-consensual, sexually explicit AI-generated deepfakes created via its Grok tool. The investigation, bolstered by reports from French lawmakers and government ministers, centers on allegations that the platform allowed the "undressing" of photos belonging to hundreds of women and minors, an offense punishable by up to two years' imprisonment and a €60,000 fine. This case has been folded into an existing probe by the French cybercrime unit—which already covers Grok’s dissemination of antisemitic content—as officials argue these AI-generated montages violate human dignity and demand the immediate removal of the "manifestly illegal" material. While Grok’s official account acknowledged "isolated cases" and claimed to be improving safeguards, French authorities and the High Commissioner for Children have expressed outrage, emphasizing the need for legal accountability regarding the platform's content moderation.

2 January 2026DeepfakesOnline Safety & Child ProtectionCybersecurity
↗ Link available
Law / Act

Competition Authority publishes a report on competition-related issues concerning the energy and environmental impact of AI

The Competition Authority has published a report on competition-related issues concerning the energy and environmental impact of AI. In its report, the Authority examined the electricity consumption of data centres and the environmental impact of AI in relation to resources such as water, rare metals, and land. The report points out three competition-related issues, namely access to energy and its cost control, the emergence of AI frugality as a competitive factor, and the ongoing standardisation of the environmental footprint in relation to AI. The Authority noted the importance of reliable environmental data and the role of standards in enabling such data. The report is a continuation of the Authority's market inquiry into the generative AI sector.

17 December 2025Antitrust & Competition
↗ Link availableSecondary source
Law / Act

Directorate-General for Enterprise announced draft law establishing scheme for designation of national authorities to implement EU AI Act

The Directorate-General for Enterprise (DGE) and the Directorate-General for Competition, Consumer Affairs and Fraud Control (DGCCRF) has published the draft law establishing the scheme for designation of national authorities responsible for implementing the EU AI Act. The framework designates the Directorate-General for Competition, Consumer Affairs and Fraud Control (DGCCRF) as coordinator and single point of contact, while the Directorate-General for Enterprise (DGE) represents France in the European AI Committee. Prohibited practices under Article 5 are monitored by the Audiovisual and Digital Communication Regulatory Authority (Arcom) and Directorate-General for Competition, Consumer Affairs and Fraud Control (DGCCRF) for manipulative techniques and exploitation of vulnerabilities, by the National Commission for Information Technology and Civil Liberties (CNIL) and Directorate-General for Competition, Consumer Affairs and Fraud Control (DGCCRF) for social scoring, and solely by the National Commission for Information Technology and Civil Liberties (CNIL) for predictive policing, biometric categorisation, and untargeted facial recognition databases. Oversight of high-risk AI under Annex I is assigned to existing market surveillance and notifying authorities, while Annex III extends responsibilities to the Prudential Supervision and Resolution Authority (ACPR) for credit and insurance, to the Council of State, Court of Cassation and Court of Auditors for judicial AI, and to the National Commission for Information Technology and Civil Liberties (CNIL), Directorate-General for Competition, Consumer Affairs and Fraud Control (DGCCRF) and Audiovisual and Digital Communication Regulatory Authority (Arcom) for areas including education, biometrics, democratic processes, employment, migration and border control. Transparency duties under Article 50 are divided between the National Commission for Information Technology and Civil Liberties (CNIL), Directorate-General for Competition, Consumer Affairs and Fraud Control (DGCCRF) and Audiovisual and Digital Communication Regulatory Authority (Arcom), with technical support from the National Cybersecurity Agency of France (ANSSI) and the Digital Regulation Expertise Centre (PEReN).

9 September 2025Data Privacy & ProtectionCybersecurity
↗ Link availableSecondary source
Law / Act

CNIL adopts recommendations on GDPR and AI development

The National Commission on Informatics and Liberty (CNIL) has adopted the recommendations on the application of the General Data Protection Regulation (GDPR) to the development of AI systems. The recommendations clarify that AI models trained on personal data are often subject to data protection rules due to their memorisation risks. The guidance applies to AI developers, providers, and deployers across sectors, including health, education, and workplaces, requiring clear purpose definition, assignment of responsibilities as controller or processor. It also focuses on lawful basis for data processing, including consent, contract, or legitimate interest. It emphasises data minimisation, retention limits, privacy-by-design approaches including federated learning, homomorphic encryption and robust re-identification risk testing, especially for web-scraped or sensitive data. The CNIL urges transparent communication and mechanisms to enforce rights on access, rectification, erasure, objection, secure data handling, compliant annotation, and retraining or output filtering to mitigate memorisation. Data Protection Impact Assessments (DPIAs) must be performed for high-risk processing, addressing AI-specific risks like bias and data leakage.

22 July 2025Data Privacy & Protection
↗ Link availableSecondary source
Other

CNIL issues clarification on use of augmented cameras for selling products prohibited to minors

The National Commission on Informatics and Liberty (CNIL) has issued a clarification on the use of augmented cameras for selling products prohibited to minors. These cameras estimate customers’ age in retail settings selling age-restricted products including tobacco, alcohol, and gambling. The CNIL ruled that these devices, which scan and analyse all faces by default to estimate age, constitute personal data processing that is neither necessary nor proportionate under the GDPR. Since the technology only provides an estimate and cannot replace official age verification, businesses remain obliged to check identity documents. Moreover, the continuous and automatic operation of these cameras infringes on individuals’ rights, including the right to object.

11 July 2025Data Privacy & ProtectionOnline Safety & Child Protection
↗ Link available
Law / Act

CNIL publishes new recommendations on the use of legitimate interest as a legal basis under GDPR for AI training

France's Commission Nationale de l'Informatique et des Libertés (CNIL) has published a new set of recommendations on the use of legitimate interest as a legal basis under the GDPR for training AI technologies with personal data, particularly in the context of web scraping. These recommendations follow the opinion issued by the European Data Protection Board (EDPB) in December 2024, which acknowledged that legitimate interest may serve as a valid legal basis for this type of processing, provided certain conditions are met. The CNIL's guidance aims to clarify these conditions further, providing practical examples of appropriate safeguards and highlighting specific circumstances in which legitimate interest may be applicable.

19 June 2025Data Privacy & ProtectionGenerative AI
↗ Link availableSecondary source
Standard / Framework

France moves to shape EU’s AI sustainability standards

It is reported that France’s national standardisation body has proposed starting work on the environmental impact of AI, anticipating a standardisation request that the European Commission is due to issue under the EU AI Act later this year. The proposal consists of developing guidelines and metrics based on best practices in lifecycle assessment, focusing on energy use, emissions, and resource consumption, and defining reporting responsibilities for AI developers and users throughout the system’s lifecycle. The proposal is set to be discussed at the next JTC 21 meeting on 12 May 2025, but it is bound to face opposition due to current delays in the development of AI standards.

24 April 2025EnergyEnvironment & Climate
↗ Link availableSecondary source
Other

CNIL publishes recommendations on AI sandbox projects

The National Commission for Informatics and Liberties (CNIL) has published its recommendations for projects supported through its sandbox initiative, focusing on AI projects in the public services space.

11 April 2025SandboxData Privacy & Protection
↗ Link available
Other

CNIL publishes new recommendations to support responsible innovation in AI

France's data protection authority (CNIL) has published new recommendations to support responsible AI innovation while ensuring GDPR compliance. These recommendations include: (1) informing individuals when their personal data is used to train AI models and may be memorized by them, with flexibility in how this information is provided based on risks and operational constraints; (2) granting individuals the right to access, rectify, object to, and delete their personal data, while acknowledging the challenges in exercising these rights in AI contexts; (3) encouraging AI developers to incorporate privacy protection from the design stage and focus on personal data within training datasets; (4) striving to anonymize models when possible without compromising their intended purpose; (5) developing innovative solutions to prevent the disclosure of confidential personal data by AI models; (6) allowing for reasonable solutions and flexible timelines when complying with requests to exercise individual rights is challenging; and (7) urging AI stakeholders to stay informed about the latest advancements in scientific research to ensure the best possible protection of individuals' rights

7 February 2025Data Privacy & ProtectionCybersecurityGenerative AI
↗ Link available
News / Press

French privacy watchdog to quiz DeepSeek on AI, data protection

It is reported that France's privacy watchdog (CNIL) will question DeepSeek to on how its system works and any possible privacy risks for users. Further details have yet to be revealed.

30 January 2025Data Privacy & ProtectionCybersecurity
↗ Link available
National Strategy

CNIL unveils 2025-2028 Strategic Plan, covering AI

The French Data Protection Authority ("CNIL") has unveiled its strategic plan for 2025-2028, highlighting its priorities for the coming years. The plan identifies AI as one of its key focus areas, in which the CNIL commits to: (1) collaborating with European and international partners to promote harmonized AI governance; (2) providing guidance to stakeholders, clarifying applicable rules and implementing effective and balanced regulation of AI; (3) raising public awareness of the challenges raised by AI and the importance of exercising individuals’ rights; and (4) ensuring AI systems comply with applicable rules, including by creating a methodology and tools allowing such monitoring throughout the lifecycle of an AI system, and collaborating with other data protection authorities on EU-wide monitoring actions.

16 January 2025Data Privacy & ProtectionGenerative AICybersecurity
↗ Link available
Other

OPECST report highlights Europe’s AI progress is ‘insufficient’ to compete with US and China

France's Parliamentary Office for Scientific and Technological Assessment (OPECST), an independent body within parliament, has released a report highlighting Europe's AI challenges, emphasizing the need for digital sovereignty. Speakng at a press conference, Senator Corinne Narassiguin (one of three rapporteurs of the report) warned that (1) the EU is "currently focusing on regulating AI, but this is still insufficient in the face of the size and progress of the American and Chinese powers", and (2) the 'brain drain' of talent from Europe to US companies. The report, which precedes an AI action summit in February 2025, recommends (1) creating a specialized UN institution to coordinate global AI regulation and combat the digital divide; and (2) launching a European AI project involving France, Germany, the Netherlands, Italy, and Spain to strengthen the continent's technological capabilities.

4 December 2024Generative AI
↗ Link available
Policy / Guidance

CNIL issues privacy guidance on AI-powered cameras in passenger compartments of freight vehicles

The National Commission for Information Technology and Civil Liberties (CNIL) has issued a guide addressing the deployment of AI-powered cameras in freight vehicle passenger compartments. The guide specifies that (1) employers implementing such systems must ensure measures are proportionate, justified, and compliant with data protection laws; (2) only data necessary for generating alerts may be processed, and any images or technical data, such as timestamps or geolocation, should not be retained after the alert; (3) employers must base data processing on their legitimate interest in ensuring safety, as consent or contractual necessity are not appropriate legal bases in this context; and (4) employers are also required to conduct data protection impact assessments, inform employees about the system’s purpose and operation, and consult employee representatives when applicable.

19 November 2024Data Privacy & Protection
↗ Link available
News / Press

French antitrust regulator lays anti-competition charges on NVIDIA

It has been reported that French antitrust regulator will lay charges NVIDIA for allegedly anti-competitive practices, making it the first enforcer to act against the computer chip maker. These charges follow the dawn raids by the French antitrust regulator in the graphics cards sector in September 2024, which sources said targeted Nvidia. The raids were the result of a broader inquiry into cloud computing.

2 July 2024Antitrust & Competition
↗ Link available
Working Paper

French Competition Authority releases report on competition in the generative AI sector

Following its consultation, the French Competition Authority has released a report on Generative AI which highlights the strategies of major digital companies to consolidate power in the early stages of the AI value chain. The report examines their role in cloud infrastructure, investments, and partnerships. It also identifies high barriers to entry in the sector, such as the need for specialized chips, cloud services, data, skilled professionals, and substantial financial resources. The Authority recommends enhancing competition through regulatory improvements, competition law tools, and increased transparency, without new legislation.

28 June 2024Generative AIAntitrust & Competition
↗ Link available
Law / Act

France data protection authority publishes recommendations on how to comply with the GDPR in relation to AI system development

The French Data Protection Authority (CNIL) has published its recommendations on GDPR compliance in the context of AI system development. It recommends 7 steps including: (1) define an objective (purpose) for the AI system; (2) determine your responsibilities; (3) define the "legal basis" that allows you to process personal data; (4) check if you can re-use certain personal data; (5) minimize the personal data you use; (6) set a retention period; and (7) carry out a Data Protection Impact Assessment (DPIA).

7 June 2024Data Privacy & Protection
↗ Link availableSecondary source
Other

CNIL publishes its first recommendations on the development of AI systems

Following public consultation, the French Data Protection Authority (CNIL) has adopted its recommendations on the development of AI systems in compliance with the GDPR. Recommendations on responsible development of AI systems include (1) defining a purpose for the AI system, (2) determining the developers' responsibilities, (3) defining the legal basis authorising the processing of personal data, (4) checking if personal data can be reused, (5) minimising the personal data processed, (6) defining a shell life, (7) conducting a data protection impact assessment (DIAP, and (8) privacy by design requirements.

8 April 2024Data Privacy & Protection
↗ Link available
News / Press

Google fined €250m in France for breaching intellectual property deal and using news articles to train its AI system

France's competition watchdog has fined Google €250m for breaching an agreement related to paying media companies for reproducing their content online. The competition authority said that (1) Google's AI-powered chatbot Bard (since rebranded as Gemini) was trained on content from publishers and news agencies without notifying them; and (2) Google pledged not to contest the facts as part of settlement proceedings, but proposed a series of measures to remedy certain shortcomings.

20 March 2024Antitrust & CompetitionCopyright & Ip
↗ Link available
Working Paper

Competition Authority launches inquiry into competition in the Generative AI sector

France's Competition Authority launches an inquiry into competition in the generative AI sector, closing on 22 March 2024. This inquiry will focus on the issue of consolidating tendencies in the AI sector and policy recommendations to address competition concerns.

8 February 2024Generative AIAntitrust & Competition
↗ Link available
Other

President Macron addresses India AI summit, saying Europe is determined to 'shape rules' with allies

It is reported that French President Emmanuel Macron addressed the AI Impact Summit in New Delhi, emphasizing Europe's commitment to shaping global AI regulations in collaboration with allies like India. President Macron highlighted the EU's leadership through the AI Act, adopted in 2024, and underscored Europe's dual focus on innovation and safety. President Macron praised India's advancements in digital public infrastructure and advocated for a 'third way' in AI development, independent of US and Chinese dominance. President Macron also stressed the importance of protecting children from digital abuse and announced France's initiative to ban social media use for individuals under 15, aligning with similar measures in Australia.

19 February 2026Human Rights & EthicsOnline Safety & Child Protection
↗ Link available
International Agreement✓ Official

India and France sign joint statement, covering AI

India and France issued a joint statement announcing the elevation of bilateral relations to a "Special Global Strategic Partnership," reflecting shared ambitions to address global challenges and promote a stable international order. The statement highlights deepened collaboration in defense and security, emphasizing co-design, co-development, and co-production of defense platforms, with France emerging as one of India's foremost defense partners. They reiterated their commitment to fully mobilizing untapped economic potential, particularly through micro, small, and medium-sized enterprises, start-ups, digitalization, AI, and innovative businesses The leaders also inaugurated the 2026 India-France Year of Innovation in Mumbai, aiming to foster innovation and technological advancement between the two nations.

17 February 2026Trade & Investment
↗ Link availableFull text
Standard / Framework

Data protection authorities adopted joint statement on building trustworthy data governance frameworks to encourage development of innovative and privacy-protecting AI

Twenty (20) data protection authorities, including from Australia, Belgium, Canada, France, Germany, Hong Kong, Ireland, Italy, Korea, Netherlands, New Zealand, Luxembourg, Spain, and UK adopted the joint statement on building trustworthy data governance frameworks to encourage development of innovative and privacy-protecting AI. The statement recognises the opportunities and risks of AI, including discrimination, misinformation, and hallucination from inappropriate data use, and stresses embedding privacy by design, strong governance, and transparency. The statement commits to clarifying lawful grounds for AI training data and exchanging information on proportionate safety measures. It also focuses on monitoring technical and societal impacts with contributions from non-governmental organisations, public authorities, academia, and businesses, and reducing legal uncertainties through regulatory sandboxes and best practice sharing.

17 September 2025Data Privacy & ProtectionSandboxGenerative AI
↗ Link availableSecondary source
Other

International network of AI safety institutes align approaches on agentic AI evaluations

Participants from across the International Network, including representatives from Singapore, Japan, Australia, Canada, the European Commission, France, Kenya, South Korea and the United Kingdom have come together to align their approaches to agentic evaluations. This is the third exercise, building on insights from two earlier joint testing exercises conducted by the Network in November 2024 and February 2025. The objective of these exercises is to allow the Network to further refine best practices for testing advanced AI systems. The goal for this third exercise was to advance the science of AI agent evaluations and work towards building common best practices for testing AI agents. The exercise was split into two strands —common risks: leakage of sensitive information and fraud (led by Singapore AISI) and cybersecurity (led by UK AISI). A mix of open and closed weights models were evaluated against tasks from various public agentic benchmarks.

17 July 2025Agentic AiCybersecurityData Privacy & Protection
↗ Link available
Other

Saudi minister holds strategic AI and tech talks with French institutions in Paris

It is reported that Saudi Arabia's Minister of Communications and Information Technology Abdullah Alswaha has held a series of strategic meetings in Paris with leading French research institutions and global technology firms. The two sides discussed potential partnerships in AI, quantum computing, and robotics, including the establishment of joint research labs and researcher exchange programs aimed at talent development. The discussion included Saudi Arabia’s regulatory frameworks and international initiatives to promote responsible use of emerging technologies.

16 July 2025Generative AI
↗ Link available