The Black Box Dilemma: Who Controls What AI Can Say? With CTGT’s Founder & CEO, Cyril Gorlla.

AI's enterprise risk isn't flashy jailbreaks but cascading errors from non-deterministic, sycophantic models lacking independent, auditable verification.

• Non-determinism is the core risk: LLMs are probabilistic next-token predictors, not deterministic functions—same query can yield different answers, creating audit/reproducibility problems for regulated firms.
• Guardrails are binary and brittle: Gorlla argues current 'generation one' guardrails just block/filter; CTGT pushes 'generation two'—surgically identifying and remediating out-of-policy outputs rather than restarting whole agentic workflows.

Themes, sentiment and reasons are AI-written paraphrases, checked so they never quote an episode. They may contain errors. Not investment advice. Episodes remain the property of their creators.