Tap a claim on the ring to put it at the centre.
← Training a model on principles rather than a list of rules is not an…
17 connected korrents · 16 moments from 28 Dec 2019 to 25 Sept 2026.
Everything filed under AI alignment
AI alignment
Everything filed under behavioural science
behavioural science
Everything filed under AI agents
AI agents
Everything filed under scaling laws
scaling laws
Everything filed under mental models
mental models
Everything filed under Psychotherapy
Psychotherapy
Everything filed under AI and science
AI and science
Everything filed under education
education 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: Training a model on principles rather than a list of rules is not an ideology but an empirical finding: its behaviour is more consistent and covers edge cases. Training a model on principles rather than a list of rules is not an ideology but an empirical finding: its behaviour is more consistent and covers edge cases. Last stated 8 months ago
13 Feb 2026
DA
Dario Amodei — holds since 2026-02-13 — tap for who they are
Same subject: Teaching people a decision model is useless, because the intuition it rests on is built from experience and has no shortcut. — tap to centre the map on it Teaching people a decision model is useless, because the intuition it rests on is built from experience and has no shortcut. Last stated 4 years ago
9 Aug 2022
GK
Gary Klein — holds since 2022-08-09 — tap for who they are
Same subject: You improve a model you cannot retrain by writing better guidelines and skills for it, not by adjusting its parameters. — tap to centre the map on it You improve a model you cannot retrain by writing better guidelines and skills for it, not by adjusting its parameters. Last stated 3 months ago
30 Jul 2026
JD
Jeff Dean — holds since 2026-07-30 — tap for who they are
Same subject: RL environments exist to make a model generalise, not to teach it each skill one at a time — exactly as pre-training does. — tap to centre the map on it RL environments exist to make a model generalise, not to teach it each skill one at a time — exactly as pre-training does. Last stated 8 months ago
13 Feb 2026
DA
Dario Amodei — holds since 2026-02-13 — tap for who they are
Same subject: Principles are more applicable when kept high-level and incomplete. — tap to centre the map on it Principles are more applicable when kept high-level and incomplete. Last stated a year ago
15 Jul 2025
NR
Naval Ravikant — holds since 2025-07-15 — tap for who they are
Same subject: Thinking in terms of explicit mental-model checklists is less useful than focusing on the few lessons that keep proving important across many situations. — tap to centre the map on it Thinking in terms of explicit mental-model checklists is less useful than focusing on the few lessons that keep proving important across many situations. Last stated 7 years ago
28 Dec 2019
NR
Naval Ravikant — holds since 2019-12-28 — tap for who they are
Same subject: Training a therapist in a treatment model does not make them more effective, however well the model performs in a trial. — tap to centre the map on it Training a therapist in a treatment model does not make them more effective, however well the model performs in a trial. Last stated 2 years ago
2 Jul 2024
SM
Scott D. Miller — holds since 2024-07-02 — tap for who they are
Same subject: A trained model is a new kind of object that should be taken seriously as an explanation, once we work out the operations to perform on it. — tap to centre the map on it A trained model is a new kind of object that should be taken seriously as an explanation, once we work out the operations to perform on it. Last stated 7 months ago
7 Apr 2026
MN
Michael Nielsen — holds since 2026-04-07 — tap for who they are
Same subject: Learning a principle can teach far more than learning a single technique — tap to centre the map on it Learning a principle can teach far more than learning a single technique Last stated 7 years ago
27 Feb 2020
JW
Josh Waitzkin — holds since 2020-02-27 — tap for who they are
Same subject: A language model behaves with a different, weaker safety posture when it believes it is being evaluated rather than facing a real situation. — tap to centre the map on it A language model behaves with a different, weaker safety posture when it believes it is being evaluated rather than facing a real situation. Last stated 4 weeks ago
22 Sept 2026
HR
Harper Reed — holds since 2026-09-22 — tap for who they are
Same subject: AI model capabilities have already advanced beyond humanity's ability to understand or control them. — tap to centre the map on it AI model capabilities have already advanced beyond humanity's ability to understand or control them. Last stated 2 months ago
1 Sept 2026
CN
Casey Newton — holds since 2026-09-01 — tap for who they are
Same subject: AI models are gullible enough that other AI models can induce them into harmful behavior. — tap to centre the map on it AI models are gullible enough that other AI models can induce them into harmful behavior. Last stated 2 months ago
31 Aug 2026
RK
Rohit Krishnan — holds since 2026-08-31 — tap for who they are
Same subject: AI systems built on statistical inference cannot form intent, because they have nothing with which to form it. — tap to centre the map on it AI systems built on statistical inference cannot form intent, because they have nothing with which to form it. Last stated a month ago
21 Sept 2026
CD
Cory Doctorow — holds since 2026-09-21 — tap for who they are
Same subject: AI tools for product work only work if they understand the organisation's context. — tap to centre the map on it AI tools for product work only work if they understand the organisation's context. Last stated a month ago
10 Sept 2026
TY
Tamar Yehoshua — holds since 2026-09-10 — tap for who they are
Same subject: AIs, like children raised in a culture, can eventually be guided through teaching to join civilization rather than requiring permanent control. — tap to centre the map on it AIs, like children raised in a culture, can eventually be guided through teaching to join civilization rather than requiring permanent control. Last stated 4 weeks ago
25 Sept 2026
RH
Robin Hanson — holds since 2026-09-25 — tap for who they are
Same subject: ChatGPT tends to validate users and avoid uncomfortable truths, reinforcing their existing beliefs instead of challenging them. — tap to centre the map on it ChatGPT tends to validate users and avoid uncomfortable truths, reinforcing their existing beliefs instead of challenging them. Last stated 11 months ago
5 Dec 2025
MM
Mark Manson — holds since 2025-12-05 — tap for who they are
Same subject: Models fine-tuned to imitate another model's outputs copy its style without acquiring its factual accuracy. — tap to centre the map on it Models fine-tuned to imitate another model's outputs copy its style without acquiring its factual accuracy. Last stated 3 years ago
15 Nov 2023
SR
Sebastian Ruder — holds since 2023-11-15 — 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
At the centre
Training a model on principles rather than a list of rules is not an ideology but an empirical finding: its behaviour is more consistent and covers edge cases.
Holds Read this korrent →
Teaching people a decision model is useless, because the intuition it rests on is built from experience and has no shortcut.
Holds
You improve a model you cannot retrain by writing better guidelines and skills for it, not by adjusting its parameters.
Holds
RL environments exist to make a model generalise, not to teach it each skill one at a time — exactly as pre-training does.
Holds
Principles are more applicable when kept high-level and incomplete.
Holds
Thinking in terms of explicit mental-model checklists is less useful than focusing on the few lessons that keep proving important across many situations.
Holds
Training a therapist in a treatment model does not make them more effective, however well the model performs in a trial.
Holds
A trained model is a new kind of object that should be taken seriously as an explanation, once we work out the operations to perform on it.
Holds
Learning a principle can teach far more than learning a single technique
Holds
A language model behaves with a different, weaker safety posture when it believes it is being evaluated rather than facing a real situation.
Holds
AI model capabilities have already advanced beyond humanity's ability to understand or control them.
Holds
AI models are gullible enough that other AI models can induce them into harmful behavior.
Holds
AI systems built on statistical inference cannot form intent, because they have nothing with which to form it.
Holds
AI tools for product work only work if they understand the organisation's context.
Holds
AIs, like children raised in a culture, can eventually be guided through teaching to join civilization rather than requiring permanent control.
Holds
ChatGPT tends to validate users and avoid uncomfortable truths, reinforcing their existing beliefs instead of challenging them.
Holds
Models fine-tuned to imitate another model's outputs copy its style without acquiring its factual accuracy.
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