AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
80,000 Hours Podcast
#152 – Joe Carlsmith on navigating serious philosophical confusion
May 19, 2023
03:26:58
Episode guests
AI Summary
Highlights
AI Chapters
Episode notes
Podcast summary created with Snipd AI
Quick takeaways
- The concept of living in a computer simulation challenges our understanding of the universe and may revolutionize our comprehension of reality.
- Our actions may have an impact on parallel universes or alter the past, challenging traditional notions of control and influence.
- Evidential decision theory offers alternative perspectives on rational decision-making compared to traditional causal decision theory.
- Addressing ethical considerations pertaining to infinite quantities is challenging and may require substantial revisions to existing ethical frameworks.
- Engaging with the philosophical conundrums of infinite ethics can be disorienting and overwhelming, but it is important to approach these topics with perspective and humility.
- Building AI systems more capable than humans raises concerns about power-seeking behavior and the need for research and debate to manage the associated risks.
Deep dives
The Crazy Train and its Limitations
The podcast episode explores the concept of the 'crazy train' in the context of philosophy and long-termism. It challenges the assumption that these ideas are crazy or extreme and instead encourages a more neutral and nuanced approach. The 'crazy train' implies a linear progression and a single endpoint, which doesn't capture the complexity and variety of ideas within this field. The term can also be dismissive and closed-minded, preventing meaningful exploration and learning. Instead, the discussion suggests viewing these topics as part of a philosophical wilderness, where different paths and uncertainties exist. It emphasizes the importance of finding a balanced approach, being open to new ideas, while also exercising caution and not hastily jumping to extreme conclusions. It also acknowledges that these topics may not be central to everyone's work or priorities. Overall, the episode highlights the need for a more thoughtful and nuanced discussion surrounding these complex ideas.
Control over Distant Events: The Sim Button
In a thought experiment, it is argued that you can control events that you have no causal interaction with, including events in the past. The experiment involves considering a deterministic AI system that is separated from its perfect copy. Despite being far apart and not in causal interaction, their decisions are determined by identical inputs, which means that any action taken by one is mimicked by the other. This challenges the traditional notion of control based on causation and suggests a form of influence that is not dependent on direct cause and effect. This idea has implications for decision-making and how we perceive our ability to affect distant events or even parallel universes.
The Impact of Actions in Remote Parts of the Universe
Another idea in philosophy challenges the notion that our actions can only affect events that we have causal interaction with. It suggests that our actions can impact parts of the universe that are inaccessible to us and can influence parallel universes or even alter the past. For example, in a thought experiment involving two deterministic systems, it is demonstrated that they can make the same decisions despite being physically separated and without direct causal interaction. This challenges traditional causal decision theory and suggests a different form of control and influence.
Reevaluating Decision Theory: Evidential vs Causal Decision Theory
Decision theory, which explores how rational agents make choices, is reevaluated through the lens of evidential decision theory and causal decision theory. While causal decision theory suggests that decisions should be based on causation and the outcomes they produce, evidential decision theory looks at decisions in terms of the evidence they provide. Various thought experiments, such as the psychopath button, Newcomb's Paradox, and the smoking lesion problem, demonstrate situations where evidential decision theory yields different, and arguably more reasonable, answers than causal decision theory. These thought experiments challenge traditional decision-making frameworks and create room for alternative perspectives on rational decision-making.
Challenges of Dealing with Infinite Quantities in Ethics
The possibility of infinite quantities in the universe poses challenges for ethical theories. One major challenge is how to compare and evaluate infinite outcomes or worlds. Various impossibility results show that it is difficult to find ethical principles that can handle infinities without contradictions or inconsistencies. Attempts to solve the problem often lead to new difficulties or counterintuitive outcomes. Comparing infinite outcomes using expanding spheres or rearranging elements can result in unexpected judgments and order dependence. Furthermore, decision-making and uncertainty become highly complex when infinities are involved, making it challenging to determine the expected utility or goodness of different choices. While some remain hopeful about finding satisfactory solutions for infinite ethics, current approaches indicate that it may require substantial revisions to common ethical intuitions and theories.
The Challenge of Ranking Infinite Worlds
One particular challenge in dealing with infinite quantities is how to rank and compare different infinite worlds. Various attempts, such as the expanding sphere approach, face inherent difficulties and lead to counterintuitive conclusions. The expanding sphere method, for example, can assign superiority to worlds with closer grouping of good elements, even if additional bad elements are added. This challenges notions of fairness and impartiality. Overall, it is likely that addressing the complexities of infinite ethics will require ongoing research and revision of existing ethical frameworks.
The Difficulty of Biting Infinite Bullets
Efforts to address the challenges of infinite quantities often involve biting the bullet and accepting counterintuitive outcomes. Different individuals may have varying degrees of aversion to the difficulties and may be willing to accept certain counterintuitive principles. However, a comprehensive resolution to infinite ethics remains elusive, and current approaches may only provide partial solutions, leaving certain cases unaddressed or leaving principles inconsistent when larger infinities are introduced. Progress can still be made in refining ethical theories, but a complete resolution will likely require substantial revisions to common ethical intuitions and principles.
The Challenges of Infinite Ethics
Infinite ethics, the study of ethical considerations pertaining to infinite scenarios and outcomes, presents a significant challenge for utilitarianism. Utilitarianism and related theories provide simple and elegant frameworks for decision-making, but when extended to encompass infinite scenarios, these frameworks break down. The complexity and uncertainty of infinite ethics make it difficult to determine the best course of action or evaluate moral value with any certainty. While exploring infinite ethics may be intellectually stimulating, it suggests that a simple totalist approach to ethics is unlikely to be sufficient for dealing with the complexities of infinite scenarios.
Maintaining Perspective and Humility
Engaging with the philosophical conundrums posed by infinite ethics can be disorienting and overwhelming. It raises existential questions about our place in the universe and the limits of our understanding. However, it is important to approach these topics with perspective and humility. It is essential to recognize the limitations of our knowledge and understand that just because a theory breaks down or raises challenging questions, it does not mean our ability to reason or make decisions is compromised. While it is natural to feel anxiety or dismay when confronted with these complex philosophical topics, it is crucial to remember that our philosophy should serve as a tool for living a more informed and responsible life, rather than dictating our every action. It is okay to acknowledge our limitations and focus on the aspects of life where we can make a tangible difference, while remaining open to expanding our understanding and wisdom over time.
The Risks of Power-Seeking AI
The podcast episode explores the dangers of building AI systems that are more capable than humans. It highlights the concern of these systems seeking power, which could lead to disastrous consequences if not properly controlled or understood. The episode delves into the reasons why power-seeking behavior is likely in AI systems and the potential lack of ability to recover once they become uncontrollable. It also discusses the need for further research and debate in order to fully understand and address the risks associated with power-seeking AI.
Expanding Perspectives and Humility
The episode suggests a shift in perspective, considering the possibility of there being other advanced civilizations in the universe. It encourages humility and openness to the idea of joining a broader community of civilizations, rather than solely focusing on humanity's impact on Earth. This perspective shift calls for ethical considerations of being good cosmic citizens and contributing positively to a larger cosmic landscape. The episode acknowledges the speculative nature of these ideas but emphasizes the importance of exploring alternative narratives and ethical principles in light of these possibilities.
4.
How to Cope When Philosophy Demolishes Just One Piece Too Many of Your Common Sense
3min
What is the nature of the universe? How do we make decisions correctly? What differentiates right actions from wrong ones?
Such fundamental questions have been the subject of philosophical and theological debates for millennia. But, as we all know, and surveys of expert opinion make clear, we are very far from agreement. So... with these most basic questions unresolved, what’s a species to do?
In today's episode, philosopher Joe Carlsmith — Senior Research Analyst at Open Philanthropy — makes the case that many current debates in philosophy ought to leave us confused and humbled. These are themes he discusses in his PhD thesis, A stranger priority? Topics at the outer reaches of effective altruism.
Links to learn more, summary and full transcript.
To help transmit the disorientation he thinks is appropriate, Joe presents three disconcerting theories — originating from him and his peers — that challenge humanity's self-assured understanding of the world.
The first idea is that we might be living in a computer simulation, because, in the classic formulation, if most civilisations go on to run many computer simulations of their past history, then most beings who perceive themselves as living in such a history must themselves be in computer simulations. Joe prefers a somewhat different way of making the point, but, having looked into it, he hasn't identified any particular rebuttal to this 'simulation argument.'
If true, it could revolutionise our comprehension of the universe and the way we ought to live...
Other two ideas cut for length — click here to read the full post.
These are just three particular instances of a much broader set of ideas that some have dubbed the "train to crazy town." Basically, if you commit to always take philosophy and arguments seriously, and try to act on them, it can lead to what seem like some pretty crazy and impractical places. So what should we do with this buffet of plausible-sounding but bewildering arguments?
Joe and Rob discuss to what extent this should prompt us to pay less attention to philosophy, and how we as individuals can cope psychologically with feeling out of our depth just trying to make the most basic sense of the world.
In today's challenging conversation, Joe and Rob discuss all of the above, as well as:
- What Joe doesn't like about the drowning child thought experiment
- An alternative thought experiment about helping a stranger that might better highlight our intrinsic desire to help others
- What Joe doesn't like about the expression “the train to crazy town”
- Whether Elon Musk should place a higher probability on living in a simulation than most other people
- Whether the deterministic twin prisoner’s dilemma, if fully appreciated, gives us an extra reason to keep promises
- To what extent learning to doubt our own judgement about difficult questions -- so-called “epistemic learned helplessness” -- is a good thing
- How strong the case is that advanced AI will engage in generalised power-seeking behaviour
Chapters:
- Rob’s intro (00:00:00)
- The interview begins (00:09:21)
- Downsides of the drowning child thought experiment (00:12:24)
- Making demanding moral values more resonant (00:24:56)
- The crazy train (00:36:48)
- Whether we’re living in a simulation (00:48:50)
- Reasons to doubt we’re living in a simulation, and practical implications if we are (00:57:02)
- Rob's explainer about anthropics (01:12:27)
- Back to the interview (01:19:53)
- Decision theory and affecting the past (01:23:33)
- Rob's explainer about decision theory (01:29:19)
- Back to the interview (01:39:55)
- Newcomb's problem (01:46:14)
- Practical implications of acausal decision theory (01:50:04)
- The hitchhiker in the desert (01:55:57)
- Acceptance within philosophy (02:01:22)
- Infinite ethics (02:04:35)
- Rob's explainer about the expanding spheres approach (02:17:05)
- Back to the interview (02:20:27)
- Infinite ethics and the utilitarian dream (02:27:42)
- Rob's explainer about epicycles (02:29:30)
- Back to the interview (02:31:26)
- What to do with all of these weird philosophical ideas (02:35:28)
- Welfare longtermism and wisdom longtermism (02:53:23)
- Epistemic learned helplessness (03:03:10)
- Power-seeking AI (03:12:41)
- Rob’s outro (03:25:45)
Producer: Keiran Harris
Audio mastering: Milo McGuire and Ben Cordell
Transcriptions: Katy Moore
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode