How Could AI Go Wrong? Near-Term Harm to P-Doom
Details
How could AI go wrong?
- This event explores the spectrum of AI risk: from near-term harms already emerging today to the deeper arguments behind misalignment and P(doom) — the probability that advanced AI could lead to catastrophic outcomes.
- We’ll use a few key ideas from Roman Yampolsky, Dario Amodei, and Demis Hassabis as starting points for a serious but accessible conversation about AI, society, and human agency.
- This is not a lecture or a networking event. It’s an in-person discussion space for people interested in technology, culture, philosophy, and the future.
- No technical background is required — just curiosity, thoughtful discussion, and a willingness to engage with difficult questions.
- Weather note: If the weather is bad, we’ll move the discussion to Fremont Brewing nearby. Any location update will be posted before the event.
Optional Videos - Roman Yampolsky: Artificial General Intelligence Will be Humanity's Last Great Invention
- Demis Hassabis & Dario Amodei: Google's Demis Hassabis, Anthropic's Dario Amodei Debate the World After AGI