Richard_Kennaway

Karma: 8,070

Computer scientist, applied mathematician. Based in the eastern part of England.

Fan of control theory in general and Perceptual Control Theory in particular. Everyone should know about these, whatever subsequent attitude to them they might reach. These, plus consciousness of abstraction dissolve a great many confusions.

I wrote the Insanity Wolf Sanity Test. There it is, work out for yourself what it means.

Change ringer since 2022. It teaches learning and grasping abstract patterns, memory, thinking with your body, thinking on your feet, fixing problems and moving on, always looking to the future and letting both the errors and successes of the past go.

I first found an LLM useful (other than for answering the question “let’s see how well the dog can walk on its hind legs”) in September 2025. As yet they do not form a regular part of anything I do.

Richard_Kennaway 5 Dec 2025 13:51 UTC
3 points
0
in reply to: keltan’s comment on: keltan’s Shortform
Claude said to keltan:

This is a criticism I’ve seen circulating in AI safety discussions

Surely Claude is bullshitting? I haven’t heard of Claude or any other AI going about reading the net in its spare time, the better to hold up its end of an informed and up to date conversation. Maybe that’s something that AIs will be doing soon, but AFAIK not yet. However, it is the sort of thing that a knowledgeable human with their finger on the pulse of events would say, so Claude says this sort of thing.

Richard_Kennaway 5 Dec 2025 13:28 UTC
0 points
0
on: Richard_Kennaway’s Shortform
Yesterday I received this email:

————
Driver & Vehicle Licensing Agency

Dear Vehicle Owner,

Our system has identified that your vehicle tax will be due on Saturday, 06 December, 2025.

To ensure that your vehicle remains legally paid and complies with applicable traffic regulations, you are required to complete the renewal process before the due date

[Update Now]

Thank you for keeping your vehicle tax up to date and for helping us maintain safe and legal roads across the UK.
————

I own a car, and the DVLA and vehicle tax are real things, so my immediate reaction to this was “Is this a genuine message?” It wasn’t, of course. But the scam email got me wondering: are the sort of people who fall for such things also the sort who think that the LLMs they talk to are people?

Richard_Kennaway 3 Dec 2025 12:34 UTC
2 points
0
in reply to: Croissanthology’s comment on: Everyone Can Be High Status In Utopia
So that’s epistemic status: drunken bullshit session.

Richard_Kennaway 2 Dec 2025 14:31 UTC
7 points
4
on: The Hidden Asymmetry in Personal Preparedness: Early Costs, Late Losses
Your analysis of early action + no disaster overlooks the fact that early action can prevent the disaster. But you never see the things that were prevented… because they were prevented. Early action only seems useful when it merely mitigates a disaster — that is, when it is not all that successful. Valiant failure is valorised over competent success.

Y2K is a case in point. There actually was a problem, and it was fixed.

Richard_Kennaway 2 Dec 2025 13:51 UTC
2 points
0
on: Richard_Kennaway’s Shortform
Here’s something that may be of interest here. It’s more EA than rationality, but I’m not on the EA forum. The Reith Lectures 2025, given by Rutger Bregman. Currently being broadcast on BBC Radio 4 and the BBC World Service, and available on BBC Sounds.

“Titled Moral Revolution, the lectures will delve into the current ‘age of immorality’, explore a growing trend for unseriousness among elites, and ask how we can follow history’s example and assemble small, committed groups to spark positive change.”

Richard_Kennaway 2 Dec 2025 8:37 UTC
2 points
−1
on: Everyone Can Be High Status In Utopia

In Rubber Souls, Bjartus Tomas argues that we can have cruely-free status games by creating underpeople without moral worth, perhaps because they are non-conscious, worth to serve as our permanent underclass.

We have these already. Factories, farms, construction sites, mines, IT departments: the sites of industry are full of them. We call them “machines”.

Richard_Kennaway 1 Dec 2025 21:19 UTC
6 points
0
on: Announcing: OpenAI’s Alignment Research Blog
Any chance of an RSS feed?

Richard_Kennaway 1 Dec 2025 14:02 UTC
2 points
0
in reply to: Ape in the coat’s comment on: The Joke

Presenting it as if they first came up with general principles which were then validated by the experience for the sake of readers comprehension, even though the casual process that led to her discovery of the theory is different.

I’m not sure what this was supposed to be. It’s a noun phrase, not a sentence.

Richard_Kennaway 30 Nov 2025 8:17 UTC
5 points
0
on: Traditional Food
Do fruit and dairy products play any role in your recommended diet?

Richard_Kennaway 28 Nov 2025 17:29 UTC
4 points
0
on: Animal welfare concerns are dominated by post-ASI futures
Isn’t everything dominated by post-ASI futures?

Richard_Kennaway 25 Nov 2025 10:28 UTC
4 points
0
on: Avoid Fooling Yourself By Believing Two Opposing Things At Once
I have two opposing beliefs about the title: it means “Don’t believe two opposing things at once — you will fool yourself”, or “Believe two opposing things at once — you will avoid fooling yourself.”

Richard_Kennaway 23 Nov 2025 19:05 UTC
2 points
0
in reply to: Mitchell_Porter’s comment on: Richard_Kennaway’s Shortform

How deep is your skepticism?

At the moment, “valence” seems to me, in the material I’ve read, no more than a name reifying a noun phrase of the form “the thing that...” It is like saying that salt tastes salty due to its saltiness, or to use a well-known example, that opium produces sleep through its dormitive principle.

Uttering a noun phrase does not conjure into existence a thing that it refers to. It can conjure an idea of such a thing into your head, and then you can look at various phenomena and “see” it there, just like you can “see” the phlogiston coming out of a burning log, or “see” the demon possessing an epileptic. That is what the whole Valence sequence looks like to me, and the valence literature that I cited.

In the context of consciousness, valence basically means the qualia of value.

My scepticism then extends to the word “value”, as applied in this context. Also, I had not got the impression from the material that valences were defined to be conscious experiences, but rather, that they pervade all decision-making in the brain, that comparison of valence is “the thing that” makes decisions. For example, the fourth reference I cited is a purely speculative article asserting the existence of “micro-valences” in low-importance rapid choices like which coffee mug to select from a cupboard. But perhaps the authors would say that these micro-valences are micro-consciously perceived.

Richard_Kennaway 22 Nov 2025 20:37 UTC
12 points
0
in reply to: Richard_Kennaway’s comment on: Dominance: The Standard Everyday Solution To Akrasia
Or to put that another way, shouldn’t the title be instead “Submission: The Standard Everyday Solution To Akrasia”? As it comes close to saying in the second-last paragraph.

Richard_Kennaway 22 Nov 2025 19:06 UTC
2 points
−8
on: Richard_Kennaway’s Shortform
What is valence, and is there such a thing?

I’m currently studying a paper that talks about the concept of “valence”, but I have not been able to discern the reasons for thinking that there is such a thing. I recalled that there was a sequence on Valence on LessWrong a few years ago. I passed it by at the time, but on reading it now, I have not found in it a reason to think that the thing is a thing. Nor am I satisfied by the references cited in the paper I am studying, which include these: 1, 2, 3, 4, 5. (Valence is not the subject of the paper, just something it touches on, so I am not expecting it to itself justify the concept beyond citing that literature.)

The hand-waving I see in the Valence sequence and in those references strikes me as similar to saying “phlogiston is the stuff that comes out of things when they burn”. ^[[1]]

What is meant by “valence”? What is the evidence that such an entity exists? How can it be measured in particular cases?
1. ↩︎
  No, retconning phlogiston as nitrogen, carbon, negative oxygen, or electrons does not work. There just isn’t anything corresponding to the concept as it was formulated in its time. No-one ever made a phlogistonometer. In fact, it was efforts to measure the things that could be measured in combustion that led to the concept being abandoned.

Richard_Kennaway 21 Nov 2025 16:49 UTC
6 points
0
on: Dominance: The Standard Everyday Solution To Akrasia
How does the Dom handle their own akrasia? Find a bigger Dom to Sub to, and so ad infinitum?

Richard_Kennaway 21 Nov 2025 16:48 UTC
7 points
2
on: Dominance: The Standard Everyday Solution To Akrasia
Doing something with another person or a group is also effective. For example, gym buddies, startup cofounders, artist groups, marriages (if done right), etc. As Alex_Altair says in another comment, it’s about accountability.

Richard_Kennaway 21 Nov 2025 10:40 UTC
7 points
1
on: Suicide Prevention Ought To Be Illegal
Have you seen “Two Arms and a Head”?

ETA: For those who haven’t, it is not a work of fiction. Also reviewed on Astral Codex Ten.

Richard_Kennaway 17 Nov 2025 10:00 UTC
3 points
0
on: A year-long journey of getting fast internet in Paris

Is France a third-world country in terms of internet service?

Always has been.

I’ve never lived there, but I remember that in the 1980s, the speed of email was primarily dependent not on distance but on the amount of government regulation of the telecoms business. Within the US, very little, so email only (!) took minutes. In the UK, not much compared with France, so email between the UK and the US would take no more than a few hours. Email with France, on the other hand, could take days and often would never arrive at all, without even a notification of failure.

Richard_Kennaway 17 Nov 2025 9:34 UTC
2 points
0
in reply to: Shankar Sivarajan’s comment on: Process Crimes and Pedantic Rules
Trumped-up process would itself be a process crime.

Richard_Kennaway 14 Nov 2025 17:06 UTC
2 points
−2
in reply to: julius vidal’s comment on: Pythia
From the perspective of this agent, i.e. me, the world and my model of it are very much distinguishable. Especially when the world surprises me by demonstrating that my model was inaccurate. Have you never discovered you were wrong about something? How can this happen if you cannot distinguish your model of the world from the world itself?