Responsible AI resources
Follow the decisions, ownership, and oversight behind a system.
Practical resources
A phased implementation sequence from initial governance and controls through a year-one programme.
Method · Agentic AI · financial-services implementation →A RACI covering the nine control families and the institution’s lines of responsibility.
Method · Agentic AI · financial-services implementation →Ensure every agent operating in or for the institution is known, uniquely identified, and lifecycle-managed.
Control · Agentic AI · financial-services implementation →Evidence and reporting items for a governing-body review of the agentic AI programme.
Method · Agentic AI · financial-services implementation →Full report: Enterprise Assurance Stack: What’s New
Published analysis
4 published briefings in this topic. Coverage reflects this collection, not the size or importance of the field.
Latest in this topic
October 8, 2026 · 5 min read · 9 cited sources
Multi-Agent Runtime Control: A theoretical study introduces Trust-Gated Capability Control (arXiv:2610.07000), deriving capability revocation guarantees for a modeled sleeper-agent compromise.
Read the briefing →August 21, 2026
6 cited sources
Anthropic raised its high-stakes misalignment risk assessment from “very low” to “low” and says its concrete AI-R&D evaluations have saturated, even though it concludes its automation threshold has not been crossed.
August 11, 2026
7 cited sources
OpenAI says it cannot rule out Critical cyber capability in an upcoming model, the first time any frontier lab has reached that determination, and has paused internal work that does not meet strengthened controls.
June 18, 2026
8 cited sources
The Fable 5 / Mythos recall has hardened into a pure eval-validity dispute, and the two sides are now reported to be negotiating a "remediate, then restore" deal.
Governance is how people decide whether, where, and how an AI system should be used. Accountability connects those decisions to named responsibilities, reviewable evidence, and a way to challenge or correct an outcome.
A written policy describes an intention. Evidence of implementation is needed to assess what happens in practice.
A visual introduction
Trace a decision from the person who owns it to the person who may need it corrected.
Explore: Ownership
Responsibility needs to attach to people with authority to approve, change, or stop the use of a system.
Follow the evidenceLook for named responsibilities and decision rights.
Responsibility needs to attach to people with authority to approve, change, or stop the use of a system.
Look for named responsibilities and decision rights.
Review should be able to question the assumptions, evidence, and consequences of a decision, with enough independence to matter.
Look for documented reasoning, meaningful challenge, and a response to findings.
Accountability includes a route to question an outcome and seek correction when something has gone wrong.
Look for an accessible appeal or correction process and evidence that it works.
Conceptual explainerSeptember 7, 2026