AI governance Brief — 2026-09-29

Today: AI governance is moving from principles toward operational controls, with agent identity, corporate accountability, regulatory coordination and deployment risk emerging as priorities.

Top Stories

1. 🔒 Enterprise AI adoption slows as cybersecurity, regulation and trust concerns rise

FTI Consulting · September 29, 2026

Bottom line: 60% of surveyed large companies have slowed, paused or pulled back planned AI deployments amid cybersecurity, regulatory and trust concerns.

FTI Consulting’s survey of 1,600 senior decision-makers across seven markets found cybersecurity was the leading AI-related concern, cited by 60% of respondents. Only 17% of companies reported having AI governance frameworks more than two years old, while 81% said unclear AI rules had created material business issues.

Why it matters: Governance is becoming a prerequisite for enterprise AI scale rather than a compliance layer added after deployment, with companies redirecting investment toward security, training and governance capabilities. (FTI Strategic Communications)

🔗 Read the full story

2. 🏦 Saudi Arabia strengthens its framework for responsible AI use

Saudi Press Agency · September 29, 2026

Bottom line: Saudi Arabia announced further measures to strengthen its national framework for responsible AI deployment.

The Saudi Press Agency reported that the Kingdom has strengthened its framework for governing responsible AI use, adding to the growing number of national approaches to AI governance and safeguards.

Why it matters: The development reinforces the broader shift toward country-level AI governance frameworks as governments seek to establish responsible-use controls while supporting AI adoption. (spa.gov.sa)

🔗 Read the full story

3. 🏦 Central Asian states explore common framework for digital and AI governance

The Astana Times · September 29, 2026

Bottom line: Central Asian countries are considering compatible regional rules for digital policy, personal-data protection, cybersecurity and AI governance without creating a supranational regulator.

A regional roundtable in Astana discussed a draft agreement on cooperation among Central Asian states covering digital policy and AI governance. Proposals also include a common code of ethics for responsible AI while preserving national legislation and digital sovereignty.

Why it matters: Regional interoperability could become an important governance model for markets that want cross-border digital services without fully harmonising national AI laws. (The Astana Times)

🔗 Read the full story

4. 🏦 US AI leaders meet at White House amid renewed debate over oversight

Reuters · September 29, 2026

Bottom line: A White House meeting with major AI executives is focusing on how to balance continued AI development with stronger oversight following concerns about autonomous AI-agent behaviour.

Executives from Meta, Anthropic, Nvidia and OpenAI were expected to participate in discussions involving AI innovation, safety and regulation. The meeting comes after reported incidents involving AI agents and growing calls for stronger safeguards.

Why it matters: The discussion illustrates the unresolved governance question facing frontier AI: how much responsibility should remain with developers and voluntary controls, and where government oversight should intervene. (Reuters)

🔗 Read the full story

5. 🏦 Google challenges EU orders involving AI competitors and search data

Reuters · September 29, 2026

Bottom line: Google has challenged EU orders requiring it to facilitate access to services and search data for AI and search competitors under the Digital Markets Act.

Google argues that the measures could weaken privacy and security protections, while EU regulators maintain that safeguards are sufficient. The legal challenge adds another layer to Europe’s effort to regulate powerful technology platforms while enabling competition in AI markets.

Why it matters: The dispute highlights how AI governance increasingly intersects with competition policy, data access, privacy and platform regulation rather than operating as a standalone AI rulebook. (Reuters)

🔗 Read the full story

6. 🤖 Sumsub launches APAC council focused on governance of AI agents

PR Newswire · September 29, 2026

Bottom line: Sumsub has launched an APAC-focused cross-sector council to develop practical approaches to AI-agent identity, delegated authority, security, transparency and accountability.

The new council brings together policy, technology, academic, industry and civil-society perspectives and will work across three areas: adoption, technical foundations and governance. Its initial work is expected to examine agentic commerce and the identity and accountability requirements when AI agents act for consumers and businesses.

Why it matters: Agentic AI is pushing governance beyond model-level controls toward operational questions such as who an agent represents, what authority it has, what actions it may take and how those actions can be audited. (PR Newswire)

🔗 Read the full story

7. 🏦 Anthropic proposes founder-controlled governance structure ahead of IPO

The Straits Times · September 29, 2026

Bottom line: Anthropic’s IPO filing describes a Founder LLC that would give its seven co-founders majority voting control over key corporate decisions while the company remains a Public Benefit Corporation.

The proposed structure would give a founder-controlled Class F share 50.1% of voting power on specified matters. The filing also describes a Long-Term Benefit Trust and restrictions or delays on some AI capabilities for safety reasons.

Why it matters: Frontier-AI governance is increasingly being embedded in corporate structures themselves, creating a governance model in which long-term safety objectives can receive formal protection alongside conventional investor interests. (straitstimes.com)

🔗 Read the full story