Sentient Index Labs Launches Independent AI Behavioral Risk Assessment

Sentient Index Labs announced in a press release the launch of S.E.B. (Sentience Evaluation Battery), a new behavioral risk assessment for AI systems. The tool measures emergent autonomy, deception, manipulation resistance, and value stability, using 58 tests across seven behavioral domains. Results are scored by four independent AI judges under blind protocols without human editorial overrides.

S.E.B. currently evaluates models from OpenAI, Anthropic, Google, xAI, and DeepSeek. The assessment produces three deliverables: AI DEFCON ratings translating raw scores into threat classifications, S-Level classifications on a ten-point sentience indicator scale, and S.E.B. Projections that track behavioral changes across model versions.

All data is secured with AES-256-GCM encryption, HMAC-SHA256 verification, and per-client watermarking. The company states it accepts no funding or sponsorship from AI vendors, ensuring independence in evaluation and maintaining client confidentiality.

S.E.B. data supports compliance documentation for frameworks such as the EU AI Act, NIST AI Risk Management Framework, OCC SR 11-7, and the FDA AI/ML SaMD framework. Evaluations are available now by subscription through the S.E.B. website.