Skip to content
IAM Insights

Explainers · Glossary

EU AI Act

European AI Act

The European Union's risk-tiered AI regulation, which for higher-risk automated decisions calls for provenance and bias-mitigation evidence an operator can produce on demand.

reference →

In our briefings

The claim each of the newest briefings on EU AI Act would have you leave with.

  1. Figuera's Sello protocol moves the record of an AI agent's actions from the agent to the services it calls. Each Sello-aware service seals a receipt body with HPKE to an owner key that the agent's JWS authorization token carries as a signed claim (owner_hpke_pk), signs the envelope with its own Ed25519 key and submits it to a public transparency log, where the owner finds it by the token's SHA-256 hash. In the paper's words, "The agent never holds any of the signing keys involved." The design assumes at least one service the agent calls is honest and uncompromised within the relevant time window: if every service colludes with the agent's operator, Figuera says, Sello cannot detect what they fabricate.

    Sello has the services an AI agent calls sign its receipts; the agent can't fake them, but it can go around themBI-FIND-2026-2275

  2. The draft puts the agent's actions in four new vCon event types: tool-call, tool-result, reasoning and system-event. Its compatibility section says vCon software that does not support the extension ignores those events and still processes the session's messages as usual, so "the agent trace is not processed". The authors call that compatible, and the "ai_agent" name is not required in the critical parameter. Those same events are where the draft asks for privacy care: tool-call inputs may contain credentials, API keys or file contents, and tool results may hold personal data. A redactor that does not support the extension cannot assess those fields and, the draft says, may reject a vCon that lists "ai_agent". The security section adds that malicious prompts kept in a record may execute if an AI system processes the record later.

    A draft rebuilds agent session records on vCon, where software without the extension ignores the tool callsBI-FIND-2026-2135

  3. The authors do not claim a working composition yet. A formal three-owner run at IETF 126 did not produce a valid result, the draft says, because its complete pre-run freeze was not issued. Revision -02 also says what it does not count. An author baseline capture reported on 27 September 2026 has "no retained raw output or environment record" and is not used as independent run evidence, a 26 September null rerun has email confirmation but no retained raw logs, and a three-computation cross-implementation row is omitted because no pinned coordinates substantiate it. The first runnable cross-slot test pack comes from co-author Iman Schrock's EMILIA Protocol: one positive four-slot case and 13 negative cases, each paired with a twin that changes only that defect and passes. That pack was built against revision -00, not -02, and the draft calls it "a candidate, not a frozen result" until a second implementation maintained by a different party runs the same bytes.

    A draft joining four agent records on one hash reports no valid joint run and now claims less about what agents didBI-FIND-2026-1894

›Cited in 31 briefings

Every briefing that discusses this term, newest first.

  1. Sello has the services an AI agent calls sign its receipts; the agent can't fake them, but it can go around them

    Paper · · arXiv · Juan Figuera · Jun 2, 2026

  2. A draft rebuilds agent session records on vCon, where software without the extension ignores the tool calls

    Draft · · IETF Datatracker · Henk Birkholz et al. · Nov 2, 2026

  3. A draft joining four agent records on one hash reports no valid joint run and now claims less about what agents did

    Draft · · IETF Datatracker · Steven Mih et al. · Aug 16, 2026

  4. Each agent and each tool call can pass its check while the system fails, an Oxford-led survey argues

    Paper · · arXiv · Christian Schroeder de Witt · Apr 29, 2026

  5. An agent audit log now names whom the agent acted for, but at low trust levels the agent may write that itself

    Draft · · IETF Datatracker · Raza Sharif, CyberSecAI · Oct 26, 2026

  6. NovaFabric seals AI agent runs as signed evidence, and its author says a signature proves a key, not a person

    Paper · · arXiv · Mohsen Seyedkazemi Ardebili · Oct 25, 2026

  7. LinkedIn, Uber and State Farm already run SPIRE for workload identity, and each plans to extend it

    Field signals · · 12 voices from the field · Nov 2024 to Oct 2026

  8. Mozilla's AI bug hunting got past hallucinated reports once a harness checked each finding mechanically

    Video · · OWASP Foundation · Uploaded October 21, 2026

  9. Barclays mandates one gateway for its agents' budgets, blocks and audit, and its AI engineering director says the setup is not yet tested at scale

    Podcast · · Insecure Agents · The Fishbowl Problem: Agent Identity Across Clouds and Vendors, with Andy McMahon (Barclays) · Oct 5, 2026

  10. The OWASP MCP Top 10 lead says few security teams can find the MCP servers developers run on their own machines

    Podcast · · The Application Security Podcast · Shadow MCP Servers: The AI Security Risk Hiding on Your Developers' Laptops · Oct 12, 2026

  11. An 'approved' flag can record four different human acts, and only a check says anything about the thing approved

    Draft · · IETF Datatracker · Andreas Ehstand · Sep 27, 2026

  12. Anthropic asks for per-agent identity; its guide leaves Claude Code sub-agent credentials unspecified

    Article · · Anthropic (Claude) · Anthropic · May 27, 2026

  13. A transparency-log profile for agents exposes edited and dropped records; an action nobody logged leaves no gap

    Draft · · IETF Datatracker · Pinar Emirdag et al. · Oct 12, 2026

  14. A post-quantum agent certificate draft warns that its own declared scopes are not proof of behaviour

    Draft · · IETF Datatracker · Reuben Burls · Jul 21, 2026

  15. Sharif's profile writes an AI agent's rights to pay and delete into an identity certificate valid up to a year

    Draft · · IETF Datatracker · Raza Sharif, CyberSecAI · Oct 2, 2026

  16. Cvent let staff build 6,000 agents to get them using AI; about 1,300 are in use, and agent owners never came up

    Podcast · · CXOTalk · Governing AI Agents at Scale: Identity, Scope, and Observability (with Glean and Cvent) | CXOTalk #914 · Mar 25, 2026

  17. A survey rates only workload identity ready for agents, and argues no standard can tell what an agent intends

    Paper · · arXiv · Takumi Otsuka et al. · Apr 25, 2026

  18. A third SCIM agent draft gives the agent five lifecycle states, a quarantine, and an archive it can never leave

    Draft · · IETF Datatracker · Saurabh Kushwaha, Oracle · Jul 27, 2026

  19. OAuth handles an agent in one company, working while you watch. The OpenID paper says that is where it stops.

    Article · · OpenID Foundation · Tobin South et al. · Oct 2025

  20. The Agent Name Service draft was rebuilt on DNS, with a GoDaddy author, and no standards body has adopted it

    Draft · · IETF Datatracker · Scott Courtney et al. · Apr 13, 2026

  21. An AI agent's refused action can be approved only by someone its mandate names, says a revised draft

    Draft · · IETF Datatracker · Lars Kersten Kroehl, CryptoKRI GmbH · May 21, 2026

  22. A preprint cuts off every sub-agent within about 40 seconds of its parent stopping, with no revocation server

    Paper · · arXiv · Saurabh Deochake, SentinelOne · May 20, 2026

  23. RSM survey discussion puts 23% at identifying digital identity as “the key”

    Podcast · · Identity at the Center · #423 - The Middle Market Identity Security Gap with Robert Snodgrass · May 18, 2026

  24. A proposed consent log for A2A and MCP records the terms an agent accepted, but the agent fills it in itself

    Paper · · arXiv · Ravi Kiran Kadaboina · Apr 16, 2026

  25. A draft format for signed AI agent session records warns a signature does not vouch for what the record says

    Draft · · IETF Datatracker · Henk Birkholz et al. · Feb 25, 2026

  26. CORAL's public agent server rejected forged messages but let a wrong key open the session stream

    Paper · · arXiv · Yedidel Louck et al. · Nov 5, 2025

  27. AAGATE maps NIST's AI risk framework onto Kubernetes, and its paper never times the 'millisecond kill switch'

    Paper · · arXiv · Ken Huang et al. · Oct 29, 2025

  28. Testing AI agents one at a time misses attacks that combine them, a Cooperative AI Foundation report argues

    Paper · · arXiv · Lewis Hammond · Feb 19, 2025

  29. An MIT-led paper has the user sign the agent's delegation token and lets OpenID Connect vouch for both parties

    Paper · · arXiv · Tobin South et al. · Jan 16, 2025

  30. The 2024 'IDs for AI' paper puts the ID on a chat session, not a model, and leaves trust in its issuer to others

    Paper · · arXiv · Alan Chan et al. · Jun 17, 2024

  31. A 2023 paper on agentic AI harms argued that calling a system an agent leaves the responsibility with people

    Paper · · arXiv · Alan Chan et al. · Feb 20, 2023

Related terms

What appears alongside EU AI Act in the same briefings.

NIST AI 100-1GDPRISO/IEC 42001DID / Verifiable CredentialsSCITTSPIRESPIFFENIST SP 800-207