Data Poisoning to Hallucinations: The Many Risks of AI Part 1

Data Poisoning to Hallucinations: The Many Risks of AI | Part 1 In this episode of Lunchtime BABLing, Dr. Shea Brown, CEO of BABL AI, is joined by Jeffery Recker for a fast-paced, unscripted deep dive into the real risks behind today’s AI systems. From data poisoning and model inversion to prompt injection, membership inference, and AI hallucinations, this lightning-round conversation breaks down the security, governance, and reliability challenges organizations must understand before deploying AI at scale. But this episode doesn’t stop at definitions. Shea and Jeffery also explore: - The difference between direct vs. indirect prompt injection - Whether AI hallucinations can ever truly be “solved” - Why AI isn’t a truth machine - Whether we’re using AI the wrong way - What responsible validation should look like in enterprise AI deployment As AI systems move from experimentation into real-world decision-making, understanding these risks isn’t optional — it’s foundational. If you're working in AI governance, assurance, compliance, risk, or deploying AI inside your organization, this conversation will help you think more critically about how these systems actually behave. 🎯 Take the FREE assessment here: https://shea-1mb3pmep.scoreapp.com/ Check out the babl.ai website for more stuff on AI Governance and Responsible AI!