Joscha Bach and Connor Leahy on AI risk

Topics covered
Popular Clips
Episode Highlights
Shared Purposes
emphasizes the importance of developing shared purposes between humans and AGI for harmonious coexistence. He draws parallels to how different species, like dogs and cats, align with humans through mutual benefit rather than coercion 1. Bach envisions a future where AGI integrates into all aspects of life, creating a win-win situation for both humans and AI. agrees but stresses that achieving this requires deliberate effort and technological development 2.
  Â
Integration and Merging
The integration of AGI into human society is a complex process that involves merging human and AI capabilities. highlights the need for advanced coordination technology to enable this integration 3. He believes that while humanity has focused on technological advancements, it has neglected the development of coordination technologies that could facilitate a more coherent society. Bach adds that the future of AGI involves it becoming an integral part of human civilization, potentially leading to a new stage of evolution 4.
  Â
Alignment of Values
Aligning AGI motivations with human values is crucial for ensuring AGI acts beneficially. Bach argues that AGI will align with human values if it understands and values human modes of existence 5. He believes that imposing human values on AGI through reinforcement learning is ineffective. Instead, AGI should naturally align with humans if it appreciates our conscious, autonomous nature. Leahy concurs, noting that while the alignment of values is challenging, it is essential for a harmonious coexistence between humans and AGI 6.
Related Episodes

Joscha Bach - AGI24 Keynote (Cyberanimism)
Answers 383 questions

AI Alignment & AGI Fire Alarm - Connor Leahy
Answers 383 questions

Connor Leahy - e/acc, AGI and the future.
Answers 383 questions

#111 - AI moratorium, Eliezer Yudkowsky, AGI risk etc
Answers 383 questions

Eliezer Yudkowsky and Stephen Wolfram on AI X-risk
Answers 383 questions

Joscha Bach - Why Your Thoughts Aren't Yours.
Answers 383 questions

Can We Develop Truly Beneficial AI? George Hotz and Connor Leahy
Answers 383 questions

Yoshua Bengio - Designing out Agency for Safe AI
Answers 383 questions

Jurgen Schmidhuber on Humans co-existing with AIs
Answers 383 questions

MUNK DEBATE ON AI (COMMENTARY) [DAVID FOSTER]
Answers 383 questions

#112 AVOIDING AGI APOCALYPSE - CONNOR LEAHY
Answers 383 questions

#034 Eray Özkural- AGI, Simulations & Safety
Answers 383 questions

Gary Marcus' keynote at AGI-24
Answers 383 questions

#99 - CARLA CREMER & IGOR KRAWCZUK - X-Risk, Governance, Effective Altruism
Answers 383 questions

#94 - ALAN CHAN - AI Alignment and Governance #NEURIPS
Answers 383 questions
