#ai-governance

Anthropic's August 5 beta routes every governed Claude Enterprise prompt to a security server you run, which returns an allow or deny verdict before inference.

Once agents ship someone owns prompts, spend, integrations and incidents. The job has at least six competing titles and a $75K to $247K+ published comp band.

A five-day plan to red-team AI agents with garak, promptfoo and PyRIT, mapped to the OWASP agentic categories and the MITRE ATLAS technique taxonomy.

The columns an agent registry needs, a 30-minute quarterly ritual, and why EU AI Act transparency is live now while high-risk duties wait until December 2027.

Five oversight patterns that keep agentic AI deployable under compliance: human gates, audit trails, scoped permissions, fallbacks, model-risk docs.

A signing-stage checklist for AI contracts: training-data clauses, retention, model deprecation, usage limits, sub-processors, exit and export rights.

Routine marketing lead scoring is generally not Annex III high-risk. A classification walkthrough, and why the deadline that matters is December 2027.

August 2, 2026 switched on GPAI enforcement powers and Article 50 transparency, while the Digital Omnibus pushed high-risk duties out to December 2027.

OpenAI, Anthropic and Google skipped NVIDIA's Open Secure AI Alliance a month after joining Akrites. That selectivity, not the absence itself, is the story.

The Pacing the Frontier letter asks for the option to slow frontier AI, not a pause. What a procurement team should actually do with that information today.

More than 1,000 AI-lab staff asked Washington for tools to pace frontier AI later — not a pause now. What the letter actually says, and who signed it.

No disclosure standard covers autonomous-agent incidents. What the CVE, NIST and EU AI Act gaps mean, and the response plan to put in place now.

A bipartisan House bill introduced July 23 would require the largest AI developers to keep a working off-switch, and would let DHS order it pulled.

OpenAI's Presence is a governed enterprise agent platform sold like consulting: no self-serve tier and no public pricing as of its July 2026 launch.

Anthropic's Deputy CISO published a four-question risk framework for agentic AI on July 17, 2026 — no product pitch. The two-mode identity model and 7 controls.

FLI's Summer 2026 index grades nine AI labs: Anthropic tops out at C+, OpenAI and Google DeepMind get C. It rates policies, not products — pair it with audits.

Codex's MultiAgentV2 now encrypts what a parent agent tells its subagents, so developers lose the local audit trail. Why the July 15 disclosure matters.

DeepMind's Demis Hassabis proposes a FINRA-style body where frontier labs submit models 30 days pre-release. Why buyers should treat it as a procurement signal.

Auto mode's classifier-gated permissions are now default on AWS Bedrock, Vertex AI, and Foundry. What enterprise teams should review before rollout.

Gartner's June 2025 prediction: over 40% of agentic AI projects canceled by end-2027. Forbes resurfaced it July 7, 2026. Why they fail and how to ship.

Claude Fable 5 didn't invent 30-day retention — it revoked zero data retention for the frontier tier. What actually changed, and how enterprises responded.

On June 26, 2026 the US gated two frontier models the same day: Anthropic's Mythos 5 and OpenAI's GPT-5.6 Sol. Inside the new managed-release era.

96% of creative orgs have AI policies and 96% of staff ignore them. Why governance fails at adoption, not drafting, and how to embed rules into the workflow.

Three days after launch, US export controls forced Anthropic to pull Fable 5 and Mythos 5 globally. The legal mechanism, the fallout, and the new vendor risk.

LLM-as-judge beats keyword filters but costs 100x more; cascades bridge the gap. A 2026 trust-and-safety guide on moderation architecture, risks, and migration.

Anthropic published binding AI safety and economic frameworks on June 10, 2026. What the compute thresholds, scope tests, and $350M pledge mean for you.

On June 9, 2026, ChatGPT's GPT-5.5 personalization reached Free and Go tiers, drawing on past chats, files, and Gmail. What agencies must do about data policy.

Asana rebrands as a coordination OS for human-agent teams after its $75M StackAI buy. What the Work Graph fixes, and which workflows to wire vs keep human.

OpenAI's Lockdown Mode severs the exfiltration stage of prompt injection, not injection entry. What it disables, who needs it, and how to deploy it RBAC-style.

VS Code now pushes agents, skills, and MCP configs to every dev via one policy file, and air-gapped BYOK drops the GitHub sign-in requirement. The rollout plan.

HITL escalation is the gap layer for production agents: evals and observability are covered, the handoff layer is not. The calibration math and async patterns.

Architectural patterns for AI data residency — region-pinned inference, prompt redaction, in-region retrieval, sovereign cloud overlays, compliance maps.

Design defensible audit trails for production agents — what to log, what to redact, retention policy, query patterns, and seven best practices inside.

Map SOC2 Trust Services Criteria to agentic AI controls — security, availability, processing integrity, confidentiality, privacy. Audit-ready framework inside.

Stand up an AI governance program in 90 days — charter, risk register, audit cadence, model-update review, ethics framework. The phased timeline inside.

Stage 8 of the agentic AI pipeline — governance charter, risk register, audit cadence, and the templates that make governance enforceable, not theatre.

Score your team's AI-assisted coding policy across 50 points — review gates, accountability, IP, redaction, secrets, and approval workflows.

200 agentic AI terms — agents, MCP, memory, planning, evaluation, governance. Examples, source links, cross-references. The reference glossary.

100 AI governance terms — EU AI Act, NIST AI RMF, ISO 42001, OECD, US state rules. Practical glossary for legal, compliance, and risk teams.

90-day enterprise agent rollout — pilot, guardrails, scale, federate phases with milestones, governance, and risk gates for IT-safe deployment.

A practical AI compliance checklist capturing every regulatory change from March 2026. Covers EU AI Act deadlines, FTC guidance, and action items.

OWASP's Agentic Top 10 defines the biggest security risks in autonomous AI systems. A plain-English guide covering each risk, real examples, and defenses.

88% of AI agents never make it to production. Root cause analysis framework with the 7 failure patterns, prevention checklist, and cost-of-failure calculator.

Congress debates preempting state AI laws with a national framework as 40+ states propose conflicting regulations. Analysis of the federal vs. state tension.

76% of organizations report unauthorized AI tool usage by employees. Shadow AI detection framework, governance policies, and risk mitigation strategies.

FTC AI policy statement deadline approaches March 11 per Trump executive order. State law preemption implications and enterprise compliance readiness checklist.

68% of small businesses use AI regularly but 77% have no formal policy. Where the ROI is real, where it's hype, and how to adopt AI without winging it.

OpenAI Frontier lets enterprises build, deploy, and manage AI agents with shared context and governance. Features, customers, and implementation guide.