Radar / AI security / Cequence Agent Trust
Cequence Agent Trust
Detects and verifies AI agent traffic on enterprise APIs — distinguishing identified agents from masquerading bots; their study found Meta's Muse agents reached more than half of the customers studied in two weeks without identifying as AI agents.
Why it matters
Names the agent-identity problem with real numbers: a large share of agentic traffic arrives undeclared, and enterprises now have a way to see it, verify it, and set policy on it.
What you could build with it
An e-commerce platform could use agent-traffic detection to separate legitimate shopping agents (which get a structured API path) from scrapers and credential-stuffing bots, cutting fraud while courting agentic buyers.
Does it hold up?
Usable and already deployed: Agent Trust is a live capability in Cequence Application and API Protection with detection rules running for every customer, backed by real traffic telemetry (Muse traffic grew ~6x in two weeks at the median studied customer; a travel customer now blocks ~1 in 5 Muse requests). Caveat: evidence comes from vendor channels; no independent buyer reviews found yet.
Built with Cequence Agent Trust
- Cequence & EMA research: 94% of enterprises trust their agents aren't over-provisioned, only 33% enforce itmea-biz.com · Coverage of the Cequence/EMA survey (46% scaling agentic AI across departments; 65% experienced an agent acting outside its intended scope, 29% with measurable business impact) — real demand evidence for Agent Trust-style controls.
- Agents Without Guardrails: The Agentic AI Governance Gap in the Enterprise (full report PDF)cequence.ai · The primary research report behind Agent Trust's launch narrative; contains the survey data on agent over-provisioning and incident rates. Found via search; PDF fetch not opened.
Learn more
First spotted on article: source.
More AI security
OpenAI, Anthropic and Google DeepMind jointly unveil cyber-focused safety models and safeguardsThe three rival labs disclosed weeks of behind-the-scenes coordination and jointly released a set of cyber-focused AI…security · JEV 0.73OpenAI textGrain watermarking for ChatGPT and Codex in the EUOpenAI will roll out its invisible textGrain watermarking system to ChatGPT/Codex users in the EU over the coming…security · JEV 0.65Cloudflare security-audit-skill: multi-phase security audits for coding agentsA coding-agent skill that runs multi-phase security audits with independently verified, machine-checkable findings.security · JEV 0.63LiveNerf: a pre-registered benchmark for post-release model driftOpen-source, pre-registered 30-day benchmark that detects whether Claude Opus 5.5 quietly gets worse after launch.security · JEV 0.61Sandlock 0.8.8: deferred commit for agent sandboxesThe process-based Linux AI sandbox (no container, no VM) ships deferred commit: every run returns a changeset of what…security · JEV 0.58OpenAI disrupts Moonshot-linked 'adversarial distillation' campaignOpenAI disclosed it shut down a coordinated campaign by 15,000+ users to extract hidden model reasoning, attributing a…security · JEV 0.58
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.