No Way Out Unlocking the Real OODA Loop: Cybernetics, AI, and Epstein's Hidden Connections
23 snips
Feb 19, 2026 Sean Manion, neuroscientist and former DoD researcher now adjunct at Duquesne, studies consciousness, TBI, and neuroscience-AI intersections. He recounts a canceled consciousness conference tied to Epstein disclosures. Conversations trace cybernetics from Wiener to Boyd’s OODA, critique engineering-only AI and LLM reliability, explore analog brain-like computing and glia, and champion small interdisciplinary dialogues for progress.
AI Snips
Chapters
Transcript
Episode notes
Combine Disciplines To Move Past AI Ceilings
- To advance AI beyond current limits, include diverse perspectives from neuroscience, cognitive science, and systems thinking.
- Combine engineering successes with deeper biological and cognitive insights rather than relying on minimal viable models.
LLMs Have Practical Limits In Trustworthy Use
- Large language models perform impressive creative and assistive tasks but show reliability limits like inconsistent retest results and hallucinations.
- Those quality gaps matter especially in high-stakes domains such as medicine and scientific research.
Use LLMs As Conversation Partners, Not Prophets
- Use LLMs as thinking partners and dialogue tools to surface ideas, not as unquestionable authorities.
- Keep humans in the loop to curate outputs and refine decisions from AI-assisted conversations.