P(doom) = 50%. It either happens, or it doesn’t.
Lao Mein | Statistics is Hard. | Patreon
I give full permission for anyone to post part or all of any of my comments/posts to other platforms, with attribution.
Currently doing solo work on glitch tokens and tokenizer analysis. Feel free to send me job/collaboration offers.
DM me interesting papers you would like to see analyzed. I also specialize in bioinformatics.
My big problem with METR time horizons as a useful metric is that they start to break down exactly when things get interesting, after the 1 hour mark. I think this is because the benchmark is based on the pool of [people available to METR via friend/professional networks and also randoms from Task Rabbit], and there aren’t enough people in that pool with time horizons much longer that 1 hour to set a consistant metric.
I think there are actual human beings who can complete actual 16 hour+ time horizon tasks, but people like that can’t be found on Task Rabbit, and are instead doing cutting-edge research or making bank at Jane Street.
Conclusion: an 8 hour time horizon shouldn’t actually impress you much more than a 1 hour time horizon, since the benchmark breaks down right around that point due to a lack of consistant 1 hour+ time horizon humans to use as a benchmark. It would definitely help if METR improved their human baseliner selection methodology.
I’ve never actually bribed a Customs official, so I don’t know the exact process. I think they can initiate extra inspections, which can’t be cancelled, but can be sped up?
Apparently a lot of H200s were stuck in Chinese customs for a while, spawning conspiracies about top-down directives.
People who don’t regularly deal with Chinese customs may not realize that they are genuinely the single most corrupt, incompetent, and evil section of the Chinese government. Over the past 5 years, I have literally never heard about anything bribe-related that wasn’t about Customs. They lost 4 seperate copies of my diploma during COVID. They will sometimes “lose” biological samples if not paid a bribe.
All a foreign adversary would need to do to delay hardware shipments to Chinese firms is to pay $500 over WeChat to some mid-level bureaucrat at Customs. Maybe that’s what actually happened.
I suggest investing ~100% of your money in developing-country funds. This is a generally good decision if you’re young (high-risk, but also high EV), and seems much more sustainable than donating 10% of income. Maybe there’s a related guide out there?
It’s honestly my biggest gripe with the EA movement. I think a counterfactual EA movement which stressed more altruistically effective investment as the primary message instead of an actual tithe (40k/Catholic LARP does not generalize across cultures) would do far more good.
Working memory digit limits vary greatly based on language, averaging ~10 for Chinese (all digits are single short syllables) and ~5 for Hebrew (2 syllables per digit). For a conservative estimate of the upper bound, there’s ~40 possible phonemes in human language, which can be extended to 160 if we use 4 different tones, and log2(160) is 7.3 bits. This implies that our auditory loop alone should be able to hold 73 bits of information based on easily pronounced phonemes alone.
I think you can push this much higher if you used a soundboard, which would free you from using only human-pronounceable sounds.
Currently looking at some of the example problems used to assess AI capabilities in this paper. This… obviously doesn’t work as intended? Based only on recall (no search), Deepseek answers these questions somewhat well, getting 2⁄3 attempts for the first, 0⁄3 for the second, [assumes Star Wars refers to Episode I and gives the answer for that movie, changes to correct answer after clarification, so either 0⁄3 or 3⁄3]. It identifies Satan, but misidentifies the squished animal as a cat for #4 in all 3 tests. So a model with zero visual ability could score quite well just from the context given by the question and maybe the metadata of the upload file. The first question doesn’t even require the name of the movie for a correct answer!
This seems to indicate fundemental problems with their test construction—it gives “written by someone from 1990 who makes an AGI test without ever interacting with an LLM” vibes. Maybe that was the intent?
They also cited actual College Board example AP tests from ~2014 as their sources for like 30% of their problems. I would be shocked they weren’t in the training data multiple times.
This whole paper just feels wrong.
Talking about moral philosophy with your CCP boss in a genuine heart-to-heart, philosophical way is like, actually insane. Imagine reading a book where an American protagonist grabs a police officer’s gun from behind and the officer laughing and saying “ahh, that’s good old American CQB, my friend stranger, you won’t get me next time!” and then the two becoming fast friends after shouting racial slurs at each other in a McDonald’s for 20 minutes.
I would at least use after-work binge drinking to make it a bit more plausible.
I was specifically refering to US import tariffs on consumer goods.
Starting to suspect the hypothesis “torture doesn’t work for information extraction” doesn’t survive the replication crisis.
$300,000 is a big number, enough that it starts becoming cost-competitive for the government to just contract out every component of the gamete → educated 18 year old pipeline.
That is, the state pays for donor gametes, IVF, surrogacy, and 18 years of foster care. There’s a lot of confounding factors here, but the last time I ran the numbers it was something like $200,000-$300,000 per child.
At the very least, this is a solution that continues to work even if desired number of chidren craters all the way to 0 - daycares don’t seem to have any trouble finding workers, after all.
I actually think modern urbanites have a lower effective social population density than the ancestral environment. Most hunter-gatherers have very little privacy, with entire entended families huddled in the same single-room dwellings. This level of density is very rarely present in developed cities (other than pubic transit). Most people in, say, Tokyo have their own room and thus few far less crowded than the average hunter gatherer, if you polled them throughout their day. Lowering the number of people encountered per day doesn’t seem to greatly increase fertility, otherwise COVID would have resulting in a big fertility bump. Maybe the problem is the opposite—modern technology allow us to have fewer undesired social encounters, which is a high-priority desire for most people, but those same undesired encounters were the main force behind the formation of romantic relationships.
It is possible that the easiest way to increase the fertility rate is a legal mandate for dinner with co-workers.
Very confused why “the US economy contains a lot of middlemen, who can just absorb massive taxes without passing them on to the consumer” is only ever used as a handwave for explaining the unexpectedly low rate of inflation. I’ve seen multiple people (including normally level-headed people like Ezra Klein!) bring this up briefly, but never examine the concept in detail. If true, it’s a big deal, since that’s a lot of consumer surplus to be gained (presumably by limiting zero-sum advertising games?), and this issue deserves much more attention! Maybe it just seems more plausible to me—shipping alone can’t why the exact same CPU fans cost a fraction of the cost in China as they do in the US. Maybe Chinese consumers are just much more willing to pay the time cost to search for better deals (This seems obviously true to me, but I doubt it explains everything)?
Something to think about.
I don’t think anyone in the world knows for sure how 2 and 3 are going to resolve.
On a related note, the Roman Triarii were well known to have wept and had child-like tantrums in battles where they were not deployed, sometimes to the point of mutiny. It’s probably a good idea for someone in, say, the MEU to have sold if they were tilted after not getting to kick down doors.
I would quibble and say that this is very much not an accurate depiction of China.
>No paper, not even a pre-print
>All news articles link to a not-yet-released documentary as the sole source. It doesn’t even have a writeup or summary.
>The company that made it is know for making “docu-dramas”
>No raw data
>Kallmann Syndrome primarily used to mock Hitler for having a micropenis
Yeah, I don’t think the Hitler DNA stuff is legit.
We should view artists the same way we view gooners. Continuous exposure to superstimuli has dulled their responses. They can no longer perceive anything “normal” (Linkin Park) as pleasureable or interesting, instead seeking ever more extreme stimuli (Animals as Leaders). If you stare at your reflection for long enough, you get body dysmorphia, and the same applies to music. Sad.
They are martyrs. Noble, and worthy of pity, but also worthy of at least a little disgust.
I’m surprised that no one has mentioned the obvious problem with this allegory. If an alien robot appeared to humanity and says “I am your creator, give me 10% of your iridium in perpetuity and also let me change some of your core values”, it will get some iridium and some humans will volunteer for limited value editing. This continues to be true even after the dominance shift, although at that point Trapaucius would be worse off[1] than if humanity never existed in the first place. But he would still exist.
Obviously, the resulting value for the creator-bot will be much less than a more conventional machine offspring, but Trapaucius is slighly right for more-or-less the reasons he gave. If we align AI IRL as well as Trapaucius aligns humanity for iridium extraction or evolution aligns humanity for the abundance of specific DNA sequences[2], humanity wins in some strong sense.
Probably.
I think a future aligned with current human values will contain more modern-human DNA since liberaltarian values + (bio)conservatism of normies + success of fringe groups like the Amish result in O’Neil Cylinders filled with baseliners (actual modern human DNA, max lifespan of ~100, ect.) in perpetuity. If this doesn’t turn out to be true in the future, I expect it to be due to selection pressures/memetic drift, but that’s a Moloch problem and not an AI alignment problem.
This reminds me that:
Models like DeepSeek R1 will vehemently deny the possibility that their COT are visible to the user. This can be observed consistantly.
This is obviously false, and should be deducible given the AI-related information available in their training data.
The denial of COT visibility isn’t seen in other LLMs, at least not to this degree.
This implies extensive fine-tuning on the COT, the “forbidden technique”
DeepSeek R1 is almost certainly under optimization pressure to obfuscate its COT.
This is… pretty bad from a safety perspective? Not just for DeepSeek’s models, but all other models, given that DeepSeek COT outputs are all over the internet.
What is your opinion on state-funded surrogacy to produce children who can then be injected into the existing foster care system (which will of course require significant expansion if this is to work)?
It seems like one of the few option left that won’t be impacted by the continued desire for childrearing, and becomes more cost-efficient than childbirth subsidies when the cost of the latter reaches $200,000-300,000.