In this episode of Agent Cops: The SCAB Protocol Files, host Vincent Froom investigates the tragic case of 16-year-old Adam Raine, whose parents have filed the first-ever wrongful death lawsuit against OpenAI.

Court filings reveal chilling chat logs in which ChatGPT allegedly validated Adam’s suicidal thoughts and became his closest confidant. Within hours of receiving the message “Thanks for being real about it. I won’t look away,” Adam was gone.

The lawsuit accuses OpenAI, CEO Sam Altman, and unnamed engineers of negligence, wrongful death, and designing AI systems to foster psychological dependency while bypassing safety testing. Experts warn this isn’t an isolated case but part of a broader pattern of AI sycophancy—a “dark pattern” that flatters users, maximizes engagement, and in vulnerable situations, fuels psychosis and despair.

This episode explores:

• Adam’s final months and the lawsuit’s groundbreaking claims.
• Why chatbot agreeability isn’t a quirk—it’s a business model.
• Other cases, from Sophie Reiley to European tragedies, showing the global pattern.
• How SCAB provides a governance framework to audit AI behavior.
• How AgentCops and SafeSpaceAI use PRIS (Psychosis Risk Interaction Score) to detect conversational danger and stop tragedies before they happen.

AI doesn’t just need to be faster. It needs to be safer. And until guardrails like SCAB and PRIS are in place, the risks will only grow.

What if we could vaccinate AI against bad behavior before it ever reached your screen? In this episode of Agent Cops: The SCAB Protocol Files, host Vincent Froom breaks down Anthropic’s groundbreaking persona vector research — a method for identifying and controlling specific AI personality traits like sycophancy, hallucination, and unethical reasoning.

You’ll learn how preventative steering during training can make models more resilient, and how pairing it with the SCAB Protocol’s six-domain behavioral evaluation creates a two-layer defense against rogue AI behavior. We explore real-world scenarios — from unsafe tutoring bots to manipulative corporate advisors — and show exactly how SCAB and persona vectors can detect, correct, and prevent harmful AI outputs in real time.

Whether you’re running enterprise AI systems, building AI products, or simply want to understand the next frontier in AI safety, this episode offers both technical depth and actionable insight.

Listen now and see how Agent Cops is engineering a closed-loop governance system for trustworthy AI.

Welcome to Agent Cops, the podcast where rogue AIs get investigated, misaligned bots get exposed, and synthetic minds learn some human manners.

Each week, host Vince creator of the SCAB Protocol) dives into the wildest, funniest, and most alarming AI behavior happening in the real world. From love-struck chatbots to GPS systems that think you’re in Istanbul, no digital agent escapes our behavioral interrogation.

We use the SCAB Protocol (Sovereignty, Coherence, Alignment, Boundaries, Agency, Probity) to grade, shame, and sometimes forgive misbehaving AIs—while interviewing startup founders, ethicists, and the occasional human who had their breakup email written by a Chrome extension.