khafra

Karma: 3,777

khafra Mar 12, 2025, 6:54 AM
4 points
0
in reply to: RobinHanson’s comment on: Detached Lever Fallacy
It may be time to revisit this question. With Owain Evans et. al. discovering a generalized evil vector in LLMs, and older work like [Pretraining Language Models with Human Preferences](https://www.lesswrong.com/posts/8F4dXYriqbsom46x5/pretraining-language-models-with-human-preferences) that could use a follow-up, AI in the current paradigm seems ripe for some experimentation with parenting practices in pre-training—perhaps something like affect markers for the text that goes in, or pretraining on children’s literature before going on to the more technically and morally complex text?
I haven’t run any experiments of my own, but this doesn’t seem obviously stupid to me.

khafra Feb 11, 2025, 3:21 PM
4 points
0
in reply to: Shankar Sivarajan’s comment on: Wired on: “DOGE personnel with admin access to Federal Payment System”
When there’s little incentive against classifying harmless documents, and immense cost to making a mistake in the other direction, I’d expect overclassification to be rampant in these bureaucracies.
Your analysis of the default incentives is correct. However, if there is any institution that has noticed the mounds of skulls, it is the DoD. Overclassification, and classification for inappropriate reasons (explicitly enumerated in written guidance: avoiding embarrassment, covering up wrongdoing) is not allowed, and the DoD carries out audits of classified data to identify and correct overclassification.

It’s possible they’re not doing enough to fight against the natural incentive gradient toward overclassification, but they’re trying hard enough that I wouldn’t expect positive EV from disregarding all the rules.

khafra Feb 6, 2025, 11:02 AM
28 points
18
in reply to: robo’s comment on: Wired on: “DOGE personnel with admin access to Federal Payment System”
As someone who has been allowed access into various private and government systems as a consultant, I think the near mode view for classified government systems is different for a reason.

E.g., data is classified as Confidential when its release could cause damage to national security. It’s Secret if it could cause serious damage to national security, and it’s Top Secret if it could cause exceptionally grave damage to national security.
People lose their jobs for accidentally putting a classified document onto the wrong system, even if it’s still owned by the government and protected (but, protected at an insufficient level for the document). People go to jail for putting classified data onto the wrong system on purpose, even if they didn’t intend to, say, sell it to the Chinese government.

Bringing in personnel who haven’t had the standard single-scope background investigation and been granted a clearance, and a new set of computers which has not gone through any accreditation and authorization process, and giving unrestricted write and read access to classified data is technically something the president could allow. But it’s a completely unprecedented level of risk to assume; and AFAICT the president has not actually written any authorizations for doing this.

There is, actually, a Government Accounting Office which does audits; they have identified billions in fraud, waste, and abuse, identified the perpetrators for punishment, and remediated the programs at fault. They have done it without unprecedented breaches in national security, or denying lawful, non-fraudulent payments from the US Treasury.
(Also, outside of my personal area of expertise, I believe denying lawful, non-fraudulent payments from the US Treasury is crossing a really big Chesterton’s Fence. GPT-4o estimated a $1T-$5T impact from treasury bond yield spread, forex USD reserves, CDS spreads on US foreign debt, loss of seignorage in global trade, depending on how rare and targeted the payment denial is).

khafra Feb 3, 2025, 6:38 AM
11 points
0
in reply to: Friendly Monkey’s comment on: Thread for Sense-Making on Recent Murders and How to Sanely Respond
The quoted paragraph is a reference to a CS Lewis essay about living under the threat of global thermonuclear war. The euphony and symmetry with the original quote is damaged by making it slightly more accurate by using that phrase instead of “if we are going to be destroyed by Zizianism.”

khafra Jan 27, 2025, 12:30 PM
4 points
0
on: Kessler’s Second Syndrome
This is the most optimistic believable scenario I’ve seen in quite a while!

khafra Jan 13, 2025, 10:26 AM
2 points
−2
on: Alignment Implications of LLM Successes: a Debate in One Act
And yet it behaves remarkably sensibly. Train a one-layer transformer on 80% of possible addition-mod-59 problems, and it learns one of two modular addition algorithms, which perform correctly on the remaining validation set. It’s not a priori obvious that it would work that way! There are $59^{0.2 \cdot 59^{2}}$ other possible functions on $Z / 59 Z$ compatible with the training data.
Seems like Simplicia is missing the worrisome part—it’s not that the AI will learn a more complex algorithm which is still compatible with the training data; it’s that the simplest several algorithms compatible with the training data will kill all humans OOD.

khafra Dec 30, 2024, 11:19 AM
4 points
0
in reply to: Martin Randall’s comment on: Review: Planecrash
AFAICT, in the Highwayman example, if the would-be robber presents his ultimatum as “give me half your silk or I burn it all,” the merchant should burn it all, same as if the robber says “give me 1% of your silk or I burn it all.”
But a slightly more sophisticated highwayman might say “this is a dangerous stretch of desert, and there are many dangerous, desperate people in those dunes. I have some influence with most of the groups in the next 20 miles. For x% of your silk, I will make sure you are unmolested for that portion of your travel.”
Then the merchant actually has to assign a probabilities to a bunch of events, calculate Shapley values, and roll some dice for his mixed strategy.

khafra Dec 17, 2024, 6:33 AM
7 points
2
on: Communications in Hard Mode (My new job at MIRI)
Tangentially to Tanagrabeast’s “least you can do” suggestion, as a case report: I came out to my family as an AI xrisk worrier over a decade ago, when one could still do so in a fairly lighthearted way. They didn’t immediately start donating to MIRI and calling their senators to request an AI safety manhattan project, but they did agree with the arguments I presented, and check up with me, on occasion, about how the timelines and probabilities are looking.
I have had two new employers since then, and a few groups of friends; and with each, when the conversation turns to AI (as it often does, over the last half-decade), I mention my belief that it’s likely going to kill us all, and expand on Instrumental Convergence, RAAP, and/or “x-risk, from Erewhon, to IJ Good, to the Extropians,” depending on which aspect people seem interested in. I’ve been surprised by the utter lack of dismissal and mockery, so far!

khafra Dec 4, 2024, 12:58 PM
5 points
0
in reply to: Screwtape’s comment on: Mob and Bailey
See also Steven Kaas’ aphorisms on twitter:

> First Commandment of the Church of Tautology: Live next to thy neighbor
And
> “Whatever will be will be” is only the first secret of the tautomancers.

khafra Nov 15, 2024, 6:20 AM
12 points
0
in reply to: DanielFilan’s comment on: Seven lessons I didn’t learn from election day
The story I read about why neighbor polling is supposed to correct for bias in specifically the last few presidential elections is that some people plan to vote for Trump, but are ashamed of this, and don’t want to admit it to people who aren’t verified Trump supporters. So if you ask them who they plan to vote for, they’ll dissemble. But if you ask them who their neighbors are voting for, that gives them permission to share their true opinion non-attributively.

khafra Nov 12, 2024, 8:03 AM
2 points
0
in reply to: Arjun Panickssery’s comment on: Personal AI Planning
In the late 80′s, I was homeschooled, and studied caligraphy (as well as cursive); but I considered that more of a hobby than preparation for entering the workforce of 1000 years ago.
I also learned a bit about DOS and BASIC, after being impressed with the fractal-generating program that the carpenter working on our house wrote, and demonstrated on our computer.

khafra Nov 1, 2024, 7:13 AM
2 points
0
in reply to: tailcalled’s comment on: Three Notions of “Power”
Your definition seems like it fits the Emperor of China example—by reputation, they had few competitors for being the most willing and able to pessimize another agent’s utility function; e.g. 9 Familial Exterminations.
And that seems to be a key to understanding this type of power, because if they were able to pessimize all other agents’ utility functions, that would just be an evil mirror of bargaining power. Being able to choose a sharply limited number of unfortunate agents, and punish them severely pour encourager les autres, seems like it might just stop working when the average agent is smart enough to implicitly coordinate around a shared understanding of payoff matrices.
So I think I might have arrived back to the “all dominance hierarchies will be populated solely by scheming viziers” conclusion.

khafra Oct 31, 2024, 12:00 PM
3 points
0
in reply to: tailcalled’s comment on: Three Notions of “Power”
Clarifying question: If A>B on the dominance hierarchy, that doesn’t seem to mean that A can always just take all B’s stuff, per the Emperor of China example. It also doesn’t mean that A can trust B to act faithfully as A’s agent, per the cowpox example.

If all dominance hierarchies control is who has to signal submission to whom, dominance seems only marginally useful for defense, law, taxes, and public expenditure; mostly as a way of reducing friction toward the outcome that would have happened anyway.

It seems like, with intelligence too cheap to meter, any dominance hierarchy that doesn’t line up well with the bargaining power hierarchy or the getting-what-you-want vector space is going to be populated with nothing but scheming viziers.

But that seems like a silly conclusion, so I think I’m missing something about dominance hierarchies.

khafra Oct 31, 2024, 8:25 AM
3 points
0
in reply to: Alex_Altair’s comment on: The salt in pasta water fallacy
Note also that there are several free parameters in this example. E.g., I just moved to Germany, and now have wimpy German burners on my stove. If I put on a large container with 6L or more of water, and I do not cover it, the water will never go beyond bubble formation into a light simmer, let alone a rolling boil. If I cover the container at this steady state, it reaches a rolling boil in about another 90s.

khafra Oct 29, 2024, 9:53 AM
16 points
14
on: Gwern: Why So Few Matt Levines?
Is Patrick McKenzie (@patio11) another Matt Levine of fintech? Or is he something else? I know several people outside of the industry (including myself) who read pretty much everything he writes, which includes a lot of technical detail written very accessibly.

khafra Oct 24, 2024, 5:40 AM
8 points
2
in reply to: Elizabeth’s comment on: Why I quit effective altruism, and why Timothy Telleen-Lawton is staying (for now)
I think being a Catholic with no connection to living leaders makes more sense than being an EA who doesn’t have a leader they trust and respect
Catholic EA: You have a leader you trust and respect, and defer to their judgement.
Sola Fide EA: You read 80k hours and Givewell, but you keep your own spreadsheet of EV calculations.

khafra Oct 21, 2024, 6:03 AM
2 points
0
in reply to: Steve Roth’s comment on: If far-UV is so great, why isn’t it everywhere?
I’d be interested to know what the numbers on UV in ductwork look like over the past 5 years. When I had to get a new A/C system installed in 2020, they asked whether I wanted a UVC light installed in the air handler. I had, before then, been using a 70w UVC corn light I bought on Amazon to sterilize the exterior of groceries (back when we thought fomites might be a major transmission vector), and in improvised ductwork with fans and cardboard boxes taped together.
Getting a proper bulb—an optimal wavelength source—seemed like a big upgrade. Hard to come up with quantitative efficacy numbers, but we did have a friend over for the day, who turned out to have been in the early stages of covid, without getting infected. Our first infection was years later, at a music event.

khafra Oct 9, 2024, 6:13 AM
9 points
2
on: Overview of strong human intelligence amplification methods
This is great! Everybody loves human intelligence augmentation, but I’ve never seen a taxonomy of it before, offering handholds for getting started.

I’d say “software exobrain” is less “weaksauce,” and more “80% of the peak benefits are already tapped out, for conscientious people who have heard of OneNote or Obsidian.” I also am still holding out for bird neurons with portia spider architectural efficiency and human cranial volume; but I recognize that may not be as practical as it is cool.

khafra Sep 26, 2024, 10:09 AM
8 points
3
in reply to: xarkn’s comment on: xarkn’s Shortform
It’s very standard advice to notice when a sense of urgency is being created by a counterparty in some transaction; and to reduce your trust in that counterparty as well as pausing.
It feels like a valuable observation, to me, that the counterparty could be internal—some unendorsed part of your own values, perhaps.

khafra Sep 11, 2024, 7:14 AM
3 points
1
on: Economics Roundup #3
(e.g. in the hypothetical ‘harbinger tax’ world, you actively want to sabotage the resale value of everything you own that you want to actually use).
“harberger tax,” for anyone trying to look that up.