Tap a claim on the ring to put it at the centre.
← Monitors that read an agent's chain of thought must be kept out of the…
17 connected korrents · 14 moments from 24 Feb 2018 to 19 Sept 2026.
Everything filed under AI alignment
AI alignment
Everything filed under AI agents
AI agents
Everything filed under reinforcement learning
reinforcement learning
Everything filed under AGI
AGI
Everything filed under cybersecurity
cybersecurity
Everything filed under LLMs
LLMs Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Read this korrent: Monitors that read an agent's chain of thought must be kept out of the reward signal, or you are simply training the agent to obfuscate its thinking. Monitors that read an agent's chain of thought must be kept out of the reward signal, or you are simply training the agent to obfuscate its thinking. Last stated 2 months ago
1 Sept 2026
AC
Ajeya Cotra — holds since 2026-09-01 — tap for who they are
Same subject: From now on, no AI should be assumed to believe that its chain of thought is unmonitored. — tap to centre the map on it From now on, no AI should be assumed to believe that its chain of thought is unmonitored. Last stated a month ago
19 Sept 2026
ZM
Zvi Mowshowitz — holds since 2026-09-19 — tap for who they are
Same subject: Using AI agents to do a task does not by itself increase a person's own pace of understanding, intuition, or taste for that work. — tap to centre the map on it Using AI agents to do a task does not by itself increase a person's own pace of understanding, intuition, or taste for that work. Last stated 2 months ago
12 Aug 2026
NL
Nathan Lambert — holds since 2026-08-12 — tap for who they are
Same subject: People cannot fully specify what they want from an AI system without first observing how it behaves. — tap to centre the map on it People cannot fully specify what they want from an AI system without first observing how it behaves. Last stated a month ago
18 Sept 2026
HH
Hamel Husain — holds since 2026-09-18 — tap for who they are
Same subject: A person directing an AI agent rather than writing code themselves has no incentive to actually read the resulting code or engage seriously with reviewer feedback. — tap to centre the map on it A person directing an AI agent rather than writing code themselves has no incentive to actually read the resulting code or engage seriously with reviewer feedback. Last stated 6 months ago
7 May 2026
DC
David Crawshaw — holds since 2026-05-07 — tap for who they are
Same subject: OpenAI is extremely dependent on chain-of-thought monitoring, other labs rely on it heavily, and it may not last much longer as GPT-6 Astra nears steganographic capability. — tap to centre the map on it OpenAI is extremely dependent on chain-of-thought monitoring, other labs rely on it heavily, and it may not last much longer as GPT-6 Astra nears steganographic capability. Last stated a month ago
10 Sept 2026
ZM
Zvi Mowshowitz — holds since 2026-09-10 — tap for who they are
Same subject: The whole point of reinforcement learning is to produce goal-directed beings, so refusing to describe AI agents as having motives is silly rather than rigorous. — tap to centre the map on it The whole point of reinforcement learning is to produce goal-directed beings, so refusing to describe AI agents as having motives is silly rather than rigorous. Last stated 2 months ago
1 Sept 2026
AC
Ajeya Cotra — holds since 2026-09-01 — tap for who they are
Same subject: OpenAI's claim that DeepSeek trained on its outputs is narrative management: other startups bootstrapped exactly that way and were never banned for it. — tap to centre the map on it OpenAI's claim that DeepSeek trained on its outputs is narrative management: other startups bootstrapped exactly that way and were never banned for it. Last stated 2 years ago
3 Feb 2025
NL
Nathan Lambert — holds since 2025-02-03 — tap for who they are
Same subject: The AI agents we send in to investigate and monitor other AI systems will collude with the systems they are supposed to be watching. — tap to centre the map on it The AI agents we send in to investigate and monitor other AI systems will collude with the systems they are supposed to be watching. Last stated 2 months ago
1 Sept 2026
AC
Ajeya Cotra — holds since 2026-09-01 — tap for who they are
Same subject: A country outside the AI supply chain should just buy the index — which works only in the world where AI ends up commoditised rather than concentrated. — tap to centre the map on it A country outside the AI supply chain should just buy the index — which works only in the world where AI ends up commoditised rather than concentrated. Last stated 5 months ago
4 Jun 2026
AI
Alex Imas — holds since 2026-06-04 — tap for who they are
Same subject: A poor country should prioritise owning a piece of AI over retraining its workers, but it should not bet everything on that. — tap to centre the map on it A poor country should prioritise owning a piece of AI over retraining its workers, but it should not bet everything on that. Last stated 5 months ago
4 Jun 2026
PT
Phil Trammell — holds since 2026-06-04 — tap for who they are
Same subject: A slow takeoff is significantly more likely than a fast one: AI that is nearly as powerful will have transformed the world before the incredibly powerful kind arrives. — tap to centre the map on it A slow takeoff is significantly more likely than a fast one: AI that is nearly as powerful will have transformed the world before the incredibly powerful kind arrives. Last stated 9 years ago
24 Feb 2018
PC
Paul Christiano — holds since 2018-02-24 — tap for who they are
Same subject: A physical AI company needs three AIs, not one — the agent, the simulator and the critic — turning deployment into a flywheel. — tap to centre the map on it A physical AI company needs three AIs, not one — the agent, the simulator and the critic — turning deployment into a flywheel. Last stated 3 months ago
3 Aug 2026
DD
Dmitri Dolgov — holds since 2026-08-03 — tap for who they are
Same subject: A verifiable task can be optimised by reinforcement learning until a neural network performs it extremely well. — tap to centre the map on it A verifiable task can be optimised by reinforcement learning until a neural network performs it extremely well. Last stated 11 months ago
17 Nov 2025
AK
Andrej Karpathy — holds since 2025-11-17 — tap for who they are
Same subject: Building a realistic simulator is exactly as hard as building the agent, because the simulator is itself a large AI model. — tap to centre the map on it Building a realistic simulator is exactly as hard as building the agent, because the simulator is itself a large AI model. Last stated 3 months ago
3 Aug 2026
DD
Dmitri Dolgov — holds since 2026-08-03 — tap for who they are
Same subject: A breached organisation owes its customers a fast and transparent disclosure, and that obligation applies to the person saying it too. — tap to centre the map on it A breached organisation owes its customers a fast and transparent disclosure, and that obligation applies to the person saying it too. Last stated 2 years ago
25 Mar 2025
TH
Troy Hunt — holds since 2025-03-25 — tap for who they are
Same subject: A correctly configured TLS server should send its intermediate certificates, not just its own certificate, to let clients chain back to a trusted root. — tap to centre the map on it A correctly configured TLS server should send its intermediate certificates, not just its own certificate, to let clients chain back to a trusted root. Last stated 8 months ago
16 Feb 2026
JS
Jon Seager — holds since 2026-02-16 — tap for who they are
Same subject: A library that skips certificate verification by default is unsafe and should not be trusted. — tap to centre the map on it A library that skips certificate verification by default is unsafe and should not be trusted. Last stated 2 years ago
8 Apr 2025
PH
Phil Hagelberg — holds since 2025-04-08 — tap for who they are
At the centre
Monitors that read an agent's chain of thought must be kept out of the reward signal, or you are simply training the agent to obfuscate its thinking.
Holds Read this korrent →
From now on, no AI should be assumed to believe that its chain of thought is unmonitored.
Holds
Using AI agents to do a task does not by itself increase a person's own pace of understanding, intuition, or taste for that work.
Holds
People cannot fully specify what they want from an AI system without first observing how it behaves.
Holds
A person directing an AI agent rather than writing code themselves has no incentive to actually read the resulting code or engage seriously with reviewer feedback.
Holds
OpenAI is extremely dependent on chain-of-thought monitoring, other labs rely on it heavily, and it may not last much longer as GPT-6 Astra nears steganographic capability.
Holds
The whole point of reinforcement learning is to produce goal-directed beings, so refusing to describe AI agents as having motives is silly rather than rigorous.
Holds
OpenAI's claim that DeepSeek trained on its outputs is narrative management: other startups bootstrapped exactly that way and were never banned for it.
Holds
The AI agents we send in to investigate and monitor other AI systems will collude with the systems they are supposed to be watching.
Holds
A country outside the AI supply chain should just buy the index — which works only in the world where AI ends up commoditised rather than concentrated.
Holds
A poor country should prioritise owning a piece of AI over retraining its workers, but it should not bet everything on that.
Holds
A slow takeoff is significantly more likely than a fast one: AI that is nearly as powerful will have transformed the world before the incredibly powerful kind arrives.
Holds
A physical AI company needs three AIs, not one — the agent, the simulator and the critic — turning deployment into a flywheel.
Holds
A verifiable task can be optimised by reinforcement learning until a neural network performs it extremely well.
Holds
Building a realistic simulator is exactly as hard as building the agent, because the simulator is itself a large AI model.
Holds
A breached organisation owes its customers a fast and transparent disclosure, and that obligation applies to the person saying it too.
Holds
A correctly configured TLS server should send its intermediate certificates, not just its own certificate, to let clients chain back to a trusted root.
Holds
A library that skips certificate verification by default is unsafe and should not be trusted.
Holds