The United Kingdom and Canada have renewed their AI security partnership with a stronger focus on interoperable standards, information sharing and coordinated frontier-model evaluations. The agreement recommits the UK AI Security Institute and Canadian AI Safety Institute to cooperation as model capabilities and associated risks develop.

The official announcement identifies three priorities. The countries will work on common AI security standards, improve information-sharing arrangements including appropriate Five Eyes channels, and expand technical cooperation through regular exchanges, possible staff placements and coordinated evaluation activity.

Interoperability is the important word. Enterprises operating across jurisdictions do not need another collection of principles that use similar language but require different evidence. They need assurance artefacts that can be reused: test results with clear scope, incident classifications, model and version identifiers, documented evaluation methods and rules for handling sensitive findings.

Mutual recognition also creates risk. A weak test should not become credible simply because two institutions accept it, and classified information-sharing must not crowd out publishable evidence for customers and researchers. The partnership should therefore define minimum evaluation quality, independence, disclosure and appeal mechanisms. For regulated organisations, the practical question is whether a model assessment completed in one country will reduce duplication in another without lowering the standard of proof. That is where bilateral cooperation could become operationally valuable.

Execution Level Governance- What audit-ready agent governance actually looks like
David Girvin, founder and CEO of Assury argues that model-in-the-loop review, AI governing AI, is fundamentally unreliable for regulated environments: even the best-performing models miss a meaningful share of violations, the reviewing model is typically provided by the same vendor being reviewed, and prompt injection or context poisoning can compromise both the acting agent and its supposed overseer simultaneously. He makes the case for deterministic, architecturally enforced controls instead, walking through Assury's approach of autonomy zones, session risk accumulation, and credential starvation, which lets a compromised agent be cut off from its tools instantly rather than relying on time-boxed access. The conversation touches on why David is sceptical of just-in-time credentialing as a solution for agent security more broadly, since agent sessions don't run on predictable human timescales, along with the current gap between how identity and security vendors are pitching agent protection and what he sees happening at the execution layer in practice. He also discusses the compliance and audit implications of probabilistic decision-making, arguing that regulated industries will increasingly need tamper-evident, hash-chained audit trails that can withstand scrutiny from auditors and regulators who are only beginning to understand agentic risk, and reflects on a named frontier lab's own published framework as an example of the gap between research and practitioner reality. Elsewhere, David reflects candidly on building a bootstrapped security company in an increasingly crowded market, why he turned down aggressive VC funding to stay in control of the product, and what a credible third-party assessment of his own gateway would need to look like given that Assury sits directly in the execution path for every customer's agents.
BrightTALK
Share this post
The link has been copied!