Alexander Gietelink Oldenziel

Karma: 6,723

″ (...) the term technical is a red flag for me, as it is many times used not for the routine business of implementing ideas but for the parts, ideas and all, which are just hard to understand and many times contain the main novelties.”
- Saharon Shelah
”A little learning is a dangerous thing ;
Drink deep, or taste not the Pierian spring” - Alexander Pope

As a true-born Dutchman I endorse Crocker’s rules.

For my most of my writing see my short-forms (new shortform, old shortform)

Twitter: @FellowHominid

Personal website: https://sites.google.com/view/afdago/home

What is the Iliad Intensive?

Leon Lang, Alexander Gietelink Oldenziel and David Udell

15 Apr 2026 18:49 UTC

74 points

2 comments2 min readLW link

Alexander Gietelink Oldenziel 13 Apr 2026 23:36 UTC
2 points
2
in reply to: Mateusz Bagiński’s comment on: Alexander Gietelink Oldenziel’s Shortform
That is true. This is not an argument [though I have tried to elucidate in some of the comments]. This is me expressing my displeasure & encouraging Habryka to reflect and reconsider.

Alexander Gietelink Oldenziel 13 Apr 2026 23:33 UTC
5 points
0
in reply to: Adele Lopez’s comment on: Alexander Gietelink Oldenziel’s Shortform
The question isn’t whether Scott Alexander is right [or virtuous]. I happen to disagree with plenty of things he is saying. That’s all beside the point. The point is that if you have the ideal to support and manage a community dedicated to honest intellectual inquiry then pushing out reasonable, polite, serious thinkers and writers is bad. Especially ones that agree with 90% of your ideology.
What is the endgame here? You think you will be able to convince world governments of a pauze if you kick out the majority of rationalists for not being pure doomer enough?

Alexander Gietelink Oldenziel 13 Apr 2026 21:10 UTC
9 points
7
in reply to: Ben Pace’s comment on: Alexander Gietelink Oldenziel’s Shortform
“Why are mommy and daddy fighting?”
I can’t escape the thought that if you are at the point of kicking out Scott Alexander you should consider if you are contributing to evaporative cooling.
Again- Scott speaks for a vast group of people within AI safety. He is not paid mouthpiece for the labs or CG or any of the powerful actors. He is ideologically very close to y’all—his views are probably more than doomer than 90% of people in AI safety broadly. I know Habryka vehemently denies this but from the outside it does very much seem like the narcissism of small differences.

Alexander Gietelink Oldenziel 13 Apr 2026 20:52 UTC
18 points
12
in reply to: Buck’s comment on: Alexander Gietelink Oldenziel’s Shortform
Of course Habryka has the right to run Lightcone as he sees fit. However, when you retaliate like the way he is considering doing with Scott Alexander—it has a real and negative impact on the epistemic health of the community. I think that is very sad and I implore Habryka to reconsider.
Scott Alexander is universally acknowledged to be a most reasonable interlocutors. His positions are very close to that of Habryka—Scott Alexander holds a doomer position that is more extreme than ~90% of the space. He has been unfailingly polite. He represents a huge swathe of the larger AI safety and rationalist space.
LessWrong serves as the public square for the AI safety community; Lightcone has the aspiration of being a lighthouse for the rationalist community, ai safety more broadly [indeed the entire lightcone]. Habryka has been a vocal advocate for strong free speech norms [rightfully so!]. He has defended the right to platform very controversial people and opinions. He has made a big deal of speaking truth to power—yet when he is the powerful party it seems he can see the mote in another but not the beam in his own eye.

Alexander Gietelink Oldenziel 13 Apr 2026 12:42 UTC
12 points
5
in reply to: Jay Bailey’s comment on: Jay Bailey’s Shortform
This is highly admirable and commendable. I am wondering if EA/AI safety community in general could encourage this norm.
- lab salaries are putting upward pressure on AI safety salaries more generally. This is crowding out charity bucks that could have been better spent elsewhere.
- We all like to pretend to ourselves that money and status isn’t that important to us. It seems that in your heart of hearts the money really did materially matter. If it matters to a moral pioneer such as yourself then it certainly will matter a lot to others.

Alexander Gietelink Oldenziel 13 Apr 2026 12:37 UTC
24 points
9
on: Alexander Gietelink Oldenziel’s Shortform
Three reasons the Iran War is good acksually.
1. The weakness of the old NATO doctrine versus drones have been further exposed. Anybody following the Ukraine war carefully will have realized we are in a military revolution and material, training, doctrine of militaries have to be completely overhauled from the ground up to compete with the Age of Drones. This was easier to dismiss when it was just a proxy power fighting Russia, but now the weakness of the US Navy is becoming much clearer.
2. Trump’s popularity with his base has been such a constant, despite a series of scandals that would have destroyed a lesser politican. The Iran War [and the Epstein files] have been the only dirt that has actually stuck with his base. Considering the strong likelihood that Trump and his ilk will try to again execute a coup in 2028 this seems to be a good thing.
3. Nuclear proliferation is bad actually. Iran is bad actually. Preventing backward states from having nukes is good. Nukes still exist. MAD works as long all participants are rational, but this becomes increasingly less likely as more and more (deranged) actors gain access to nuclear weapons.

Alexander Gietelink Oldenziel 13 Apr 2026 12:14 UTC
17 points
−52
on: Alexander Gietelink Oldenziel’s Shortform
The owner of this website seems to be willing to entertain petty and adversarial actions against very reasonable people he disagrees with.
Original: https://x.com/ohabryka/status/2043543756977430593
more context: https://x.com/slatestarcodex/status/2042329870076637242
I am quite dismayed by Habryka’s words and ways and I hope he will reconsider them. LessWrong has been without doubt the greatest project for good to have appeared from the greater EA-sphere and indeed- the world. Having witnessed the birth of LW 2.0 I can say with some embarrassment that it was not at all clear that anything like LessWrong could have existed. It is a testimony to Habryka’s [and Ben Pace and the rest of the team!] industry, derring-do, and ingenuity that we are able to convene here at all.
It is a deep shame that Lightcone funding is still precarious. It is a shame that Habryka and his team have been blacklisted from Coefficient Giving funding.
Habryka has been in the forefront on defending free speech norms, often taking principled stances in favour of people expressing unpopular opinions. In a personal capacity, I have worked with Oliver on several occasions and came away with a sense that he was highly competent and morally courageous.
That is all to say, I say this with love, care and a deep appreciation of Habryka & Lightcone’s work, morals and achievements.
Dear Habryka, you have shown great courage and eloquence in hectoring powerful figures and organizations in public, occasionally to the severe financial detriment of you and your organisation. You have been a tireless advocate, in speech and action, that speaking truth to power is a moral duty. So allow me to to follow your example and write this publicly. This is not the way. I hope you reflect and reconsider.

Alexander Gietelink Oldenziel 13 Apr 2026 11:28 UTC
9 points
0
on: Alexander Gietelink Oldenziel’s Shortform
Occasionally I hear arguments that ‘scaling will lead to AGI’ or that ‘compute & data limits will prevent the step to AGI’. Or ‘if LLMs knows so much why aren’t they automating every job already’.
This seems straightforwardly false to me and predicated on a conceptual confusion.
A human can be a general intelligence. The compute power of the human brain is probably roughly equivalent to 0.1-10 H100s.
The step to general intelligence is algorithmic not scaling resources. Yes, more compute will yield greater capabilities but fundamentally a dangerous and superintelligent AGI could probably run on your laptop.
As a helpful analogy, think about the difference between a Universal Turing machine and a finite state machine. Larger finite state machine can recognize more complex languages, so scaling a finite state machine will make it more powerful- however there will always be computational tasks that a universal Turing machine can do that a large finite state machine can’t.

Announcing ILIADIII: AENEID

Eddie Jean, windows, Alexander Gietelink Oldenziel and David Udell

11 Apr 2026 3:05 UTC

32 points

0 comments1 min readLW link

Alexander Gietelink Oldenziel 31 Mar 2026 16:44 UTC
4 points
0
in reply to: niplav’s comment on: Take note of how brightness makes you feel
Niplav, true heir of gwern

Alexander Gietelink Oldenziel 29 Mar 2026 22:48 UTC
6 points
0
in reply to: Adam Scholl’s comment on: adam_scholl’s Shortform
I strongly agree. The whole focus on banning AI data centers seems somewhat myopic & misguided

Alexander Gietelink Oldenziel 28 Mar 2026 8:01 UTC
9 points
0
on: The AIXI perspective on AI Safety
As a fellow fan of AIXI, and more generally idealized models of agency and the alignment problem I’m happy to see this comparison. Imho AIXI- and AIXI-adjacent thinking is still a tiny minority, even for more mathematically inclined people in alignment. That’s a shame.
I like your comparison of AIXI with other theory frameworks. People should make inside-view comparisons like this more often.
That said- my main objection to AIXI for alignment is that it is blind to inner alignment. The model of AIXI already assumes you encode the agent with whatever utility function you want. That limits it application to alignment directly.
I am most excited for modern learning theory & RL theory to meet AIXI & agent foundations. I think, at least mathematically, this is where the alignment problem fundamentally lies.

Alexander Gietelink Oldenziel 27 Mar 2026 15:16 UTC
2 points
0
in reply to: leogao’s comment on: leogao’s Shortform
always had you as von stauffenberg in my head :)

Alexander Gietelink Oldenziel 24 Mar 2026 8:37 UTC
2 points
0
in reply to: Anastasia Wei’s comment on: Announcement: Iliad Intensive + Iliad Fellowship
Yes sufficient! Claude can fill in the rest =)

Alexander Gietelink Oldenziel 22 Mar 2026 7:41 UTC
10 points
6
on: China Derangement Syndrome
How China fights its bordering neighbour.

Alexander Gietelink Oldenziel 14 Mar 2026 15:24 UTC
LW: 2 AF: 1
0
AF
on: Steelmanning heuristic arguments
>>> Basically, a very optimistic hope for understanding “why circuits behave differently from functions” is to make statements that, conditional on some “surprising” property P holding for a circuit (here “surprising” means that it occurs much more for circuits than functions), we can deduce that, with high probability, the circuit internals have some property A(C).

My understanding of the NCP-conjecture is different [and I claim more interesting].
If we want we can go from a surprising property Q [your P] holding for the function to some property A(C)… the canonical way is to assume some probability model and do Bayesian updating.
In other words A(C) will be some averaged property over the Bayesian posterior. Assuming everything is discrete and finite [we work with Boolean circuit of fixed size] and a uniform prior this is all completely explicit.

The more interesting thing to me is the following.

Let’s say Q is a disjunction Q = Q_1 OR Q_2
In that case, informally we could reason: well Q is the case because either Q_1 is the case or Q_2 is the case. Crucially the property A(C) holding for the underlying circuit C may depend on whether we have Q_1 or Q_2.
Just like we can have many different proofs for the truth of the same statements more generally we can have many different heuristic arguments for the same ‘truish’ proposition. The Bayesian posterior almagates and averages over all of them.

-------
Random Speculations:

Let me speculate on what an heuristic argument may be in this case.
One proposal for what a heuristic argument would be is the following:
parameterize probability distributions on the finite discrete space of Boolean circuits and Boolean functions by an exponential family [iiuc this can be done using log-linear models]. Then consider the outrageous statement Q(C) and consider those distributions that make it very likely.

In other words if we parameterize our distributions by w then P(Q(C)|w) is assumed high.
For a given C and a specific w we may ask is P(C|w) high or even equal to 1?
In that case, that w and the associated sufficient statistics may be regarded as a reason for Q(C).

of course—there are many such w potentially. We need w that are not too complicated to describe for each possible circuit C.
Let’s look at the fiber F=F^{-1}(Q)=\{C\in Circuit| Q(C)\}.
Now the fiber F breaks up in components F_i—intuitively there are different classes of fuctions/circuits that can give rise to the property Q. We’re looking for simple descriptions / sufficient statistics w such that of a given component F_i is assigned high probability by w.

How do we find these components F_i—if they even make sense? One way is to recursively break up the fiber of F of Q=Q(C) by logically decomposing Q and using subformulas to describe boolean features that may fix a sufficient statistic for above.

Alexander Gietelink Oldenziel 9 Mar 2026 14:23 UTC
14 points
2
in reply to: Daniel Murfet’s comment on: Cole Wyeth’s Shortform
For those of us who couldn’t make it to Sydney last year—what was your previous position on Solomonoff induction and how did it change as a result of talking with Cole in Sydney?

Alexander Gietelink Oldenziel 6 Mar 2026 17:05 UTC
4 points
0
in reply to: Cleo Nardo’s comment on: strawberry calm’s Shortform
By the way you might also be interested in Aristotle & Boethius’s original work on the topic.

Alexander Gietelink Oldenziel 6 Mar 2026 17:04 UTC
5 points
0
in reply to: Cleo Nardo’s comment on: strawberry calm’s Shortform
What you are talking about are ′ generic objects’ .
They (re)appear in language/linguistics, logic and computer science. E.g. in computer science they are closely related to ‘parametric polymorphisms’/ ′ generic data types’, in logic with Girard’s System F.
There is a large literature on them. Eg semantics of generic objects is occasionally used to semanticize anaphora, see e.g. here
The most interesting bit to me is that they are related to a dual connective to universal/existential quantifiers that are defined by elimination.
Concretely—if I have to prove \exists \phi(x), the logical introduction rule says that I have to supply an instance t such that \phi(t). I can then conclude \exist \phi(t) by the \exists-introduction rule. In a standard natural deduction style type theory by logical harmony there is a corresponding elimination rule that would eliminate these \exists \phi(x).

However, we could also posit a new quantifier that is defined by it’s elimination rule. So in that case \forall x \phi(x) would produce a c such that \phi(c) iff \forall x \phi(x), in other words a generic object c.
It’s a complete non-starter. To see what kind of object every-dog is, you can examine it’s properties. For any predicate φ, every-dog satisfies φ iff every dog is φ. For example, every-dog is a mammal, weighs less than 800 tonnes, etc. But every-dog lacks the property of being four-legged, and lacks the property of not-four-legged, since dogs vary on this. Every-dog violates excluded middle — it’s properties are gappy.
exactly. so this generic c would be your ‘every-dog’. Since we’re working in type theory we are not forced to use excluded-middle so these objects can happily exist without contradiction. =)
Some-dog is the dual: it has every property instantiated by at least one dog. So it’s simultaneously brown-all-over and white-all-over, male and female, three months old and twelve years old. Some-dog violates non-contradiction — it’s properties are glutty.
No-dog has every property that no dog has. It’s a prime number. It’s the Eiffel Tower.
No-dog is not so interesting for non-typed settings. However, if we type or use various universes we can do much richer things.
rk. More generally in full (Martin-Lof) type theory
We can extend from \exists and \forall to all \Sigma and \Pi -types [and Universe types]. This would give constructions of arbitrary generic objects and universes satisfying any collection of properties...

Despite being a terrible semantics of quantifiers, this corresponds (if you squint) to the isomorphism between a finite-dimensional vector space and its double dual. Think of the domain of objects D as a vector space, and predicates as linear functionals on D — elements of D*. Then quantifiers live in D**, functionals on predicates. There’s a canonical map D → D** given by x ↦ (f ↦ f(x)): each object corresponds to the quantifier “evaluate at x”, i.e. the proper name quantifier. In finite dimensions this map is an isomorphism — every quantifier is a name in disguise.
Yes you are definitely on to something here. There are some connections with double-dual constructions. I think the definite treatment is still left to be worked out though.

Alexander Gietelink Oldenziel

What is the Iliad In­ten­sive?

An­nounc­ing ILIADIII: AENEID

What is the Iliad Intensive?

Announcing ILIADIII: AENEID