Steve Byrnes

What if tomorrow’s AI learned and acted more like a brain? Physicist-turned-safety researcher Steve Byrnes explores how neuroscience might guide safer AGI, from built-in drives and learning systems to the challenge of keeping powerful machines aligned. The conversation also ventures into AI consciousness, digital sentience, and whether brain-like systems could arrive within decades.

Join guests Bogdan Ionut-Cirstea, Steve Byrnes, Gunnar Zarnacke, Jack Foxabbott, and Seong Hah Cho, who contribute critical insights on AI alignment. They discuss an intriguing concept called self-other overlap, which aims to optimize AI models by aligning their reasoning about themselves and others. Early experiments suggest this technique can reduce deceptive behaviors in AI. With its scalable nature and minimal need for interpretability, self-other overlap could be a game-changer in creating pro-social AI.