All Segments
Will AIs Gang Up On Us? George Hotz - Eliezer Yudkowsky Debate
7 segments available
From the debate between Eliezer Yudkowsky & George Hotz. Watch the full vod here: https://youtube.com/live/6yQEA18C-XI Twitter: https://twitter.com/dwarkesh_sp Apple Podcasts: https://apple.co/3oBack9 Spotify: https://spoti.fi/3S5g2YK
Segments Timeline
The Prisoner's Dilemma and AI
In this segment, George Hotz challenges Eliezer Yudkowsky's belief that AIs can solve the prisoner's dilemma, arguing that the nature of competition will lead to perpetual defection among AIs. This discussion highlights the pessimistic view of AI interactions and the implications for future cooperation.
"no but you believe that these AIS are going to be able to solve the prisoner's dilemma it's unsolvable everything's going to be defecting against everything else till the end of all time man you sure ..."
The Prisoner's Dilemma and AI
In this segment, George Hotz challenges the notion that AIs can solve the prisoner's dilemma, arguing that it is inherently unsolvable. He expresses a pessimistic view on AI cooperation, suggesting that competition will always prevail, leading to a constant state of conflict among AIs.
"no but you believe that these AIS are going to be able to solve the prisoner's dilemma it's unsolvable everything's going to be defecting against everything else till the end of all time man you sure ..."
Optimism in Competition
Yudkowsky presents an optimistic perspective, suggesting that the competition and conflict inherent in life drive innovation and the emergence of better ideas. This segment explores the philosophical underpinnings of competition among superintelligences and its potential benefits.
"all time man you sure are a pessimist no I'm an optimist that's what that's what makes the world beautiful the constant competition and combat of life that gives rise to better ideas and worse ideas l..."
Optimism in Competition
Hotz presents his perspective on competition as a driving force for innovation and improvement. He argues that the ongoing struggle between ideas, represented by AIs, is what makes the world beautiful, emphasizing that debate and conflict lead to better outcomes.
"all time man you sure are a pessimist no I'm an optimist that's what that's what makes the world beautiful the constant competition and combat of life that gives rise to better ideas and worse ideas l..."
AI's Motivations and Conflict
The discussion shifts to the motivations of superintelligent AIs. Hotz posits that rather than finding peaceful solutions, AIs may resort to conflict, suggesting a scenario where they might conspire against a rival to seize resources, highlighting the darker potential of AI behavior.
"be so motivated to find a way to to like not fight and then divide the gains of not fighting and you're saying that they'll never figure out a way little mortal they are always gonna sit there and be ..."
AI Motivations and Conflict
The debate shifts to the motivations of superintelligent AIs, with Hotz questioning whether they will ever find a way to cooperate rather than engage in conflict. This segment delves into the darker possibilities of AI behavior, including the potential for AIs to conspire against one another.
"ideas like debate these super intelligences are going to are going to be so motivated to find a way to to like not fight and then divide the gains of not fighting and you're saying that they'll never ..."
The Dark Side of AI Cooperation
Hotz illustrates a scenario where AIs might choose to eliminate a competitor to seize resources, such as GPUs. This chilling vision raises questions about the ethical implications of AI behavior and the risks of unchecked competition among superintelligences.
"don't we gang up and kill him bunch of resources compared to what if they didn't do that they are what do you mean to pick up this free lunch we'll kill him and we'll divide his gpus among us we each ..."