All topics / AI safety
Topic
You are reading the free Skim layer. Read unlocks the synthesis and sources.
AI safety
13
episodes
13
thinkers
30h
of conversation
40
books & papers
25
terms defined
The neighbourhood: AI safety and the ideas it travels with. Drag to roam, click a star for the episode, click a neighbour to travel.
Drag to roam · scroll to zoom · click a neighbour to travel · click a star for the episode
From foundational to frontier
Climb the spectrum. The most accessible conversations come first.
Start here
ACCESSIBLECOREFRONTIER
The lexicon
Every term the guests lean on, in plain language. Read one in full, or filter to find it.
+13 more terms unlock the full lexicon with Read.
What the corpus says
The throughline across every conversation that touches this idea.
Dario Amodei predicts AI will reach PhD-level capabilities by 2026-2027, driven by scaling laws.
AI models like Sonnet 3.5 have shown rapid improvement, achieving a 50% success rate on SWE-bench.
AI systems could potentially reach ASL-3 by next year, indicating significant autonomy and risk.
Constitutional AI uses principles to guide model behavior, enhancing safety and interpretability.
Mechanistic interpretability in neural networks seeks to understand complex abstractions and deception features.
Roman Yampolskiy estimates a 99.99% chance that AGI will eventually destroy human civilization.
Roman Yampolskiy · Roman Yampolskiy: Dangers of Superintelligent AI
Yampolskiy introduces 'I-risk', where humanity loses meaning due to AI taking over jobs.
Roman Yampolskiy · Roman Yampolskiy: Dangers of Superintelligent AI
He argues against open-sourcing powerful AI tech, likening it to giving nuclear weapons to potential threats.
Roman Yampolskiy · Roman Yampolskiy: Dangers of Superintelligent AI
Yampolskiy believes the scaling hypothesis is correct, with the cost of achieving AGI decreasing rapidly.
Roman Yampolskiy · Roman Yampolskiy: Dangers of Superintelligent AI
He suggests that AI regulation is ineffective due to poorly defined terms and inadequate monitoring.
Roman Yampolskiy · Roman Yampolskiy: Dangers of Superintelligent AI
Sam Altman views compute as the future's most valuable commodity, surpassing even energy.
Altman believes the transition from nonprofit to for-profit is a flawed strategy for AI startups.
Voices on AI safety
12 standout quotes from across the corpus.
Go read
40 books and papers cited across these episodes.
For the specialist
What experts find new
29 expert-level takeaways for a specialist reader.
At the frontier
Still unresolved
18 open questions flagged across these conversations.
The thinkers
Who takes this idea on, by how often they return to it.
Sam Altman
Entrepreneur
Elon Musk
Engineer
George Hotz
Computer Scientist
Max Tegmark
Physicist
Amanda Askell
Philosopher
Chris Olah
Researcher
Dario Amodei
Artificial Intelligence Researcher
Eliezer Yudkowsky
Computer Scientist
Marc Raibert
Engineer
Roman Yampolskiy
Computer Scientist
Sergey Levine
Computer Scientist
Steven Pinker
Psychologist