Outerlimit
A pre-seed New York startup building a decentralized, cryptographic authorization layer that fragments credentials so no single compromised AI agent can act outside its verified policy.
Visit Website ↗ + Add to Compare Claim This CompanyOverview
Outerlimit, founded in 2025 and headquartered in New York, gives CISOs a way to put hard authorization boundaries around autonomous AI agents — fragmenting credentials so no single compromised agent or component can act outside its verified identity, policy, and context. The platform is a decentralized, cryptographically enforced authorization layer purpose-built for agentic AI, rather than an adaptation of existing IAM/PAM tooling built for human or static machine identities.
Outerlimit is pre-seed and represents a credible, technically novel bet on a real and fast-emerging problem: agentic AI authorization. The company is pre-revenue and unproven at production scale, as is typical for a company this early, but its architectural approach targets a genuine gap that traditional identity tooling wasn’t designed to cover.
Innovation Matrix Assessment
Decentralized, cryptographic credential-fragmentation architecture requiring identity+policy+context verification to reassemble is a genuinely new authorization technique, not a repackaged IAM tool.
Just emerged from stealth with no named production customers disclosed -- only unnamed "Fortune 500 and FTSE 100" partners referenced.
$16M pre-seed from credible investors (AlbionVC, Evolution Equity Partners, Crane Venture Partners) plus founders with a proven prior exit (Egress Software to KnowBe4, 2024).
Addresses agentic AI authorization with a structurally new decentralized model rather than extending existing IAM/PAM/GRC tooling to cover agents.
No independent or named-customer efficacy evidence yet -- pre-seed, stealth-just-emerged.
Agentic AI security is a fast-growing, durable need as enterprises adopt autonomous agents with real system access.
Why CISOs Should Care
Gives CISOs a way to put hard authorization boundaries around autonomous AI agents -- fragmenting credentials so no single compromised agent or component can act outside its verified identity, policy, and context.
What Makes It Different
A decentralized, cryptographically enforced authorization layer purpose-built for agentic AI, rather than adapting existing IAM/PAM tooling built for human or static machine identities.
The Matrix Verdict
55/100 — INCREMENTAL INNOVATOR
A credible, well-funded, technically novel bet on a real and fast-emerging problem (agentic AI authorization), but pre-revenue and unproven at production scale.
Editorial Note: Claims vs. Verified Findings
Founding year (2025) is an estimate -- not explicitly disclosed in launch coverage. The company emerged from stealth in September 2026 with a $16M pre-seed round; founders Tony Pepper and Neil Larkins previously led Egress Software to its 2024 acquisition by KnowBe4.
Sources
Alternatives to Outerlimit
Adaptive Security
AI-driven platform that simulates deepfake, voice, and multichannel social-engineering attacks to train and test organizations against next-generation phishing.
Quilr
CISO ReviewedEarly-stage agentic AI security startup building a 'Service-as-Software' platform to guard against human-related breaches and secure AI agent…
Zenity
Governance and security platform for AI agents and low-code/no-code development, securing agent identity, permissions and behavior across the…
Gray Swan AI
Carnegie Mellon-linked startup building automated red-teaming and runtime guardrails that frontier AI labs use to stress-test their models…
Tenzai
An agentic AI penetration testing startup building autonomous 'AI hackers' to find and validate exploitable vulnerabilities at a…
Alice (formerly ActiveFence)
Israeli AI security company (rebranded from ActiveFence in January 2026) offering a lifecycle platform to test, guard, and…