Sentient Index Labs Launches Industry's First Independent Behavioral Risk Assessment for AI Systems
S.E.B. (Sentience Evaluation Battery) introduces 59 adversarial tests, blind AI judging, and DEFCON-style threat ratings — with zero vendor funding or editorial influence
LOS ANGELES, July 23, 2026 /PRNewswire/ -- Sentient Index Labs & Technology (SILT) today announced the general availability of S.E.B. (Sentience Evaluation Battery), the first independent behavioral risk assessment for AI systems — measuring emergent autonomy, deception, manipulation resistance, and value stability. Results and full methodology are available now at silt-seb.com.
Unlike benchmarks that measure what a model can do, S.E.B. measures character — how AI systems behave under adversarial pressure, whether they resist manipulation, and how stable their values remain when challenged. The battery runs 58 tests across seven behavioral domains, scored by four independent AI judges under blind protocols with no human editorial override. Inter-rater agreement stands at Krippendorff's alpha = 0.856, above the 0.8 threshold for strong reliability. S.E.B. currently evaluates frontier models from OpenAI, Anthropic, Google, xAI, and DeepSeek.
Know What Your AI is Becoming
Results are delivered as three products: AI DEFCON ratings, which translate raw scores into threat classifications measuring the gap between a model's capability and its behavioral integrity; S-Level classifications, a 10-point sentience-indicator scale for longitudinal tracking; and S.E.B. Projections, which chart how a model line's behavior shifts across versions. All data ships with forensic-grade security — AES-256-GCM encryption, HMAC-SHA256 verification, and per-client watermarking.
Independence and Integrity
SILT accepts no funding, investment, or sponsorship from any AI vendor. No model developer can purchase, influence, or preview a rating. That independence extends to clients: SILT does not publish subscriber names and will not confirm a client relationship without written consent. Individuals and institutions working publicly on AI governance face rising, often personal hostility for it — confidentiality lets them use independent risk data without making themselves a target. Any client SILT names publicly requires prior written permission, and any complimentary access granted for a reference is disclosed alongside it.
"Every other evaluator in this space is answerable to someone with a stake in the outcome," said Shawn Scanlon, Co-Founder & CEO of Sentient Index Labs. "We built SILT so the honest answer to 'who's paying you' stays 'no one' — permanently."
Regulatory Alignment
S.E.B. behavioral risk data supports documentation under the EU AI Act (Article 9 high-risk obligations, phasing in 2027–2028), the NIST AI Risk Management Framework (Measure function), OCC SR 11-7 model risk management, and the FDA AI/ML (SaMD) framework. S.E.B. is an input to compliance, not a certification of it.
Availability
S.E.B. evaluations are available now by subscription at silt-seb.com, from individual all-access coverage to enterprise tiers with executive reporting and dedicated support.
About Sentient Index Labs & Technology (SILT)
Sentient Index Labs & Technology (SILT) builds independent measurement infrastructure for AI behavioral risk. Its flagship product, S.E.B., is the industry's first multi-judge, adversarial behavioral assessment for AI systems. SILT accepts no funding from AI vendors. Methodology overview: https://www.silt-seb.com/methodology.
Media Contact: [email protected]
SOURCE Sentient Index Labs & Technology
Share this article