AIFl@sh is a monthly collection of critical upgrades and changes across the GCC’s AI landscape – with a smattering of relevant changes from outside of the region. Read this month’s newsletter on LinkedIn – or click here to subscribe!

GCC

Kingdom of Saudi Arabia

SDAIA – National AI risk management framework launched

On 14 July 2026, SDAIA launched the national AI risk management framework, providing government and private-sector entities with a unified methodology for identifying, assessing, treating and continuously monitoring AI risks. The framework recognises that AI risks may emerge unexpectedly, change over time and be difficult to explain or reproduce, and structures risk management around four phases: defining context and scope, identifying and assessing risks, treating risks and continuous monitoring and review.

Saudi Arabia is translating responsible AI principles into practical lifecycle controls. Organisations should align AI inventories and approval processes with documented risk assessments, treatment plans, monitoring, escalation and clear accountability for AI-related risks.

The United Arab Emirates

Agentic AI implementation plans reviewed

On 1 July 2026, the Ministerial Council for AI and Development reviewed the timeframe, objectives, implementation stages and execution tracks for adopting agentic AI across the federal government. Proposed projects cover government business operations, corporate governance, cybersecurity and other areas as the UAE progresses its plan to transition government services and operations towards AI-enabled models.

The UAE is moving from announcing its agentic AI ambition to establishing a coordinated implementation programme. Organisations adopting AI agents should define authority limits, human approval points, access permissions, cybersecurity safeguards, audit trails, fallback procedures and accountability for actions performed by AI systems.

Kingdom of Bahrain

Bahrain participates in inaugural global AI governance dialogue

On 8 July 2026, Bahrain’s Minister of Transportation and Telecommunications participated in the first United Nations Global Dialogue on AI Governance in Geneva. Bahrain highlighted its commitment to expanding the use of AI across sectors to improve government services and operational efficiency, while emphasising international cooperation, transparency, trust and inclusiveness as essential to the safe and responsible use of AI.

While this does not introduce a new binding requirement, it reinforces Bahrain’s implementation-led and internationally-aligned AI governance direction. Organisations should be looking translate the outcomes of the dialogue into practical controls covering AI procurement, data governance, transparency, accountability and workforce awareness.

State of Qatar

Shura Council extends AI governance review

On 6 July 2026, Qatar’s Shura Council approved a request to extend the work of its Health, General Services and the Environment Committee to continue studying a proposal for a general debate on the regulation and governance of AI in Qatar.

While no AI-specific legislation has yet been adopted, the extension confirms that legislative consideration remains active. Organisations operating in Qatar should continue strengthening controls around privacy, data governance, ethical use and accountability, while monitoring the committee’s future findings and recommendations.

Sultanate of Oman

Oman becomes a founding member of the World AI Cooperation Organisation

On 17 July 2026, Oman became one of 29 founding members of the World AI Cooperation Organisation after signing its establishing agreement in Shanghai. The independent intergovernmental organisation aims to strengthen international cooperation and global AI governance, promote safe and beneficial AI and help bridge differences in AI capabilities between countries.

Membership does not on its own introduce domestic regulatory requirements, but it positions Oman to contribute to international AI governance discussions and benefit from member-state experience, research and technical cooperation. Organisations should monitor how this participation influences Oman’s future national standards, capacity-building programmes and international AI partnerships.

State of Kuwait

Kuwait participates in inaugural global AI governance dialogue

On 8 July 2026, Kuwait announced its participation in the first United Nations Global Dialogue on AI Governance and associated events in Geneva. The delegation included representatives from the Ministry of Foreign Affairs, the Communications and Information Technology Regulatory Authority and the Central Agency for Information Technology, with delegates participating in discussions covering AI governance, digital infrastructure, digital policy and capacity building.

The multi-agency delegation indicates that Kuwait is approaching AI governance as a coordinated national issue connected to digital transformation and infrastructure policy. Organisations should expect future governance expectations to combine innovation with transparency, accountability, secure infrastructure and international good practice.

International

European Union

AI omnibus enters into force | Article 50 guidance is finalised

On 20 July 2026, the European Commission published final guidelines on the AI Act’s Article 50 transparency obligations, which apply from 2 August 2026 to certain AI systems, including requirements concerning disclosure of AI interaction and the marking or labelling of AI-generated and manipulated content. On 27 July 2026, the AI Omnibus entered into force, extending the application of requirements for Annex III high-risk AI systems to 2 December 2027 and for high-risk AI systems embedded in Annex I regulated products to 2 August 2028.

The extended high-risk deadlines do not postpone the Article 50 transparency requirements. Organisations should therefore already be separating compliance workstreams and implementing applicable user notices, machine-readable marking and content-labelling measures, while continuing to classify high-risk systems and prepare the documentation, testing and governance evidence required by the (now delayed) deadlines.

China

Rules for human-like AI interaction services take effect

On 15 July 2026, China’s interim measures for the administration of anthropomorphic AI interaction services took effect. The measures prohibit providers from offering virtual relatives, partners or equivalent virtual intimate relationships to minors; require parental or guardian consent for other covered services provided to children under 14; and require safeguards relating to emotional dependence, excessive use and user safety. Providers must also intervene in serious situations involving self-harm or threats to life and contact a guardian or emergency contact where required.

China is extending AI governance beyond content and data risks into emotional, psychological and behavioural safety. Providers of AI companions and other human-like interaction services should implement age assurance, guardian-consent mechanisms, usage controls, distress detection, emergency escalation and clear boundaries between AI systems and human relationships.

United Kingdom

FCA publishes review of AI in retail financial services

On 6 July 2026, the UK Financial Conduct Authority published the Mills Review on how AI may reshape retail financial services, consumers, markets and regulation through 2030. The review recommends that the FCA assess within three to six months whether the existing regulatory perimeter adequately captures the influence of general-purpose AI tools on financial decisions, and consider how accountability and consumer protection should apply as AI systems become increasingly autonomous.

The review indicates that the UK’s sector-led approach may require closer oversight of AI tools operating outside traditional regulated institutions but influencing regulated outcomes. Financial-services organisations should assess third-party model dependencies, advice and guidance boundaries, operational resilience, consumer recourse and accountability for AI-mediated decisions.

United States of America

FTC consults on AI accuracy and consumer expectations

On 1 July 2026, the Federal Trade Commission sought public comment on a proposed policy statement concerning the suppression of accuracy in AI systems. The proposal considers whether manipulating AI outputs contrary to reasonable consumer expectations of objectivity and accuracy, including for undisclosed ideological objectives, may amount to unfair or deceptive conduct under Section 5 of the FTC Act.

The proposal illustrates how existing consumer-protection law may be applied to AI design, behaviour and marketing claims. AI providers should be able to substantiate representations concerning accuracy, objectivity and suitability and should document material system instructions, tuning decisions and interventions that may affect outputs presented to consumers.

Canada

Public consultation on AI transparency launched

On 23 July 2026, the Government of Canada launched a public consultation on increasing transparency for AI systems and AI-generated content, with submissions open until 23 September 2026. The consultation seeks views on where transparency is most important and what actions the government should consider to support safe, responsible and reliable AI.

Transparency is emerging as a central feature of Canada’s evolving AI policy approach. Organisations should review whether users are clearly informed when they interact with AI, how AI-generated outputs are identified, and whether adequate documentation, provenance information and explanations are available for material AI-assisted decisions.

United Nations

First global dialogue on AI governance convened

On 1 July 2026, the United Nations Independent International Scientific Panel on AI published its preliminary report on AI opportunities, risks and impacts. The report informed the inaugural global dialogue on AI governance, held in Geneva on 6-7 July 2026 and involving governments, the private sector, civil society, academia and the technical community. At the opening, the UN Secretary-General also called for an AI child safety pledge centred on child-specific safety testing, independent oversight and accountability.

Global AI governance is increasingly supported by recurring scientific assessment and a formal multistakeholder dialogue rather than relying on voluntary principles. Organisations should expect child safety, inclusion, independent assurance and cross-border accountability to become increasingly prominent governance themes.

AI governance is expanding beyond privacy, bias, safety and model risk into sustainability and environmental accountability. Organisations should begin considering compute-related environmental impact, supplier energy sources, data-centre dependencies and sustainability disclosures as part of responsible AI governance.