ACX
Book Review: What We Owe The Future
Scott Alexander reviews Will MacAskill's 'What We Owe The Future', a book arguing for longtermism and considering our moral obligations to future generations.
50 min read 7,661 words
A thread through the archive
AI safety
Essays tagged “AI safety”. Follow an idea across the archive.
Essays on AI safety
79 essaysACX
Why Not Slow AI Progress?
Scott examines why the AI safety community isn't more actively opposing AI development, exploring the complex dynamics between AI capabilities and safety efforts.
20 min read 3,004 words
ACX
ELK And The Problem Of Truthful AI
Scott Alexander examines the Eliciting Latent Knowledge (ELK) problem in AI alignment and various proposed solutions.
42 min read 6,490 words
ACX
Deceptively Aligned Mesa-Optimizers: It's Not Funny If I Have To Explain It
Scott Alexander explains mesa-optimizers in AI alignment, their potential risks, and the challenges of creating truly aligned AI systems.
23 min read 3,483 words
ACX
Information Markets, Decision Markets, Attention Markets, Action Markets
Scott Alexander explores different types of prediction markets, their uses, limitations, and ethical considerations in decision-making and information gathering.
20 min read 3,064 words
ACX
Practically-A-Book Review: Yudkowsky Contra Ngo On Agents
Scott Alexander reviews a dialogue between Yudkowsky and Ngo on AI alignment difficulty, exploring the challenges of creating safe superintelligent AI.
33 min read 5,013 words
ACX
Links For December
Scott Alexander shares 34 diverse and interesting links on topics ranging from historical curiosities to current scientific debates, with brief commentaries on each.
17 min read 2,607 words
ACX
Links For October
A compilation of 30 diverse links on various topics including science, history, technology, and culture, with brief commentaries by Scott Alexander.
19 min read 2,850 words
ACX
Highlights From The Comments On Acemoglu And AI
Scott Alexander responds to comments on his AI risk post, discussing AI self-awareness, narrow vs. general AI, catastrophe probabilities, and research priorities.
37 min read 5,613 words
ACX
Updated Look At Long-Term AI Risks
Scott Alexander discusses a new expert survey on long-term AI risks, highlighting the diverse scenarios considered and the lack of consensus on specific threats.
9 min read 1,322 words
SSC
Book Review: Human Compatible
Stuart Russell's 'Human Compatible' presents AI safety concerns and potential solutions in an accessible way, though the reviewer has reservations about its treatment of current AI issues.
33 min read 5,043 words
SSC
A Maximally Lazy Guide To Giving To Charity In 2019
Scott Alexander offers a guide for lazy people to donate to effective charities, ranging from easy options like EA Funds to more specific recommendations in various cause areas.
8 min read 1,213 words
SSC
Book Review: Reframing Superintelligence
Scott reviews 'Reframing Superintelligence' by Eric Drexler, which proposes future AI as specialized services rather than general agents, contrasting with Nick Bostrom's scenarios.
16 min read 2,371 words
SSC
Links 4/18: Siter Plate
Scott shares a diverse collection of interesting links and news items, covering topics from ancient history to modern politics and technology, with brief commentary on each.
15 min read 2,179 words
SSC
The Hour I First Believed
Scott Alexander speculates on how concepts from decision theory and AI could lead to the emergence of a God-like entity across the multiverse, which judges and potentially rewards human behavior.
18 min read 2,790 words
SSC
SSC Journal Club: AI Timelines
Scott analyzes a new survey of AI researchers, showing diverse opinions on AI timelines and risks, with many acknowledging potential dangers but few prioritizing safety research.
16 min read 2,467 words
SSC
Notes From The Asilomar Conference On Beneficial AI
Scott Alexander shares his experiences and insights from the Asilomar Conference on Beneficial AI, covering various aspects of AI development, risks, and ethical considerations.
19 min read 2,933 words
LW
Superintelligence FAQ
Scott explains why superintelligent AI poses serious risks, how a 'fast takeoff' could leave insufficient time to solve safety problems, and why even simple-sounding goals could lead to catastrophic outcomes without proper alignment research.
54 min read 8,281 words
SSC
Links 9/16: URL Of The Chaldees
A links post featuring stories about AI research, scientific studies, historical oddities, and current events, ranging from antibacterials in soap to disco riots at baseball games.
10 min read 1,531 words
SSC
Links 7/16: Peter Linklage
A collection of interesting links covering various topics from scientific studies, social issues, and current events, including problems with brain imaging studies, AI safety developments, and surprising research findings.
13 min read 1,933 words
SSC
Ascended Economy?
Scott Alexander speculates on a future 'ascended economy' where AI-run corporations dominate, discussing potential implications and risks of such a system.
19 min read 2,884 words
SSC
Should AI Be Open?
Scott Alexander argues that OpenAI's open-source strategy for AI development could be dangerous, potentially risking human extinction if AI progresses rapidly.
28 min read 4,288 words
SSC
Links 7/15: From Link, Where Rocky Peaks Climb Bleak And Bare
A collection of interesting links covering various topics from scientific studies and economic policies to cultural oddities, with particular focus on deworming program controversies and workplace discrimination research.
13 min read 1,865 words
SSC
Links 7/15: Link-Carbon Battery
A diverse collection of links and studies covering education, race, healthcare, and various other topics, with Scott's commentary and analysis on each item.
16 min read 2,412 words
SSC
No Time Like The Present For AI Safety Work
Scott argues for the importance of starting AI safety research now, presenting key problems and reasons why early work is crucial.
31 min read 4,792 words