Revision History
AI Agents · 22 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Google confirmed (after WSJ report) rather than disclosed
Version 22 · Sep 28, 2026, 01:50 PM
- Update: add 2026 agent sandbox-escape disclosures and NVIDIA Open Agent Safety Platform
Version 21 · Sep 28, 2026, 01:46 PM
- Bring the AI agents survey current through August 2026: add protocols and governance (MCP 2026-07-28 stateless rewrite, A2A, AGENTS.md, AG-UI), a frameworks and harnesses section with current status, benchmark contamination and reward hacking in agent evaluation, the confused-deputy security framing with documented incidents, and published deployment evidence
Version 20 · Aug 1, 2026, 11:15 AM
- Correction: Repaired 28 bibliography emphasis spans that rendered as literal Markdown markers; factual prose, citations, source text, URLs, internal links, and categories are unchanged.
Version 19 · Jul 30, 2026, 10:44 AM
- Correction: replace market forecasts, product and framework catalogues, volatile rankings, malformed references, and unsupported autonomy claims with a primary-source systems account of agent scope, architecture, planning, tools, memory, multi-agent coordination, evaluation, security, oversight, and reproducible deployment.
Version 18 · Jul 28, 2026, 05:16 PM