@Anthony Honestly, I'm considering moving up from a range centered on 2025 - PaLM came close on Winogrande from recollection. At this point it looks increasingly likely to resolve within the next few years. I might go so far as to say that if there was a real push they might be able to deliberately do this within the next year.

I'll say this, I think people are over certain here. Even midterm results are not particularly predictive of presidential election results. Attempting to predict divergences from the base rate this severe this far out is dubious. I'll give the community a nod of a few points given it appears they disagree with me, but I'm sticking close to the base rate until more data is available.

This comment was originally posted on Joe Biden

No idea if this was intended as a prediction or what, but I thought it was interesting.

Minerva author on AI solving math: - IMO gold by 2026 seems reasonable

I'm one of the people who are substantially above the community - and while this rocketing to the top of the front page was part of why I predicted on it, it isn't all. A significant portion is that I dislike predicting on extreme improbabilities (this despite thinking that X-risk probabilities are useful for long term prediction building), and a belief that the community is underestimating the risk here. Essentially, my argument is two fold - the first is in regard to domestic politics. One thing that's stood out clear the last few months is Putin's at...

So, this is closing soon and I'm still not clear as to whether the IT restriction at the border counts for resolution. Dependent on whether it does, this is either already fulfilled, but waiting for sufficient credible media reports that may not exist or occur, or unlikely to happen.

Well, I'm either incorrect or about to mess this up for myself given I predicted this only had a 10% chance of happening (something I have no idea why I did, given I was much higher back when I first predicted), but looking at fiscal results of 2022 it appears outlays did fall more than this from 2021. [JOINT STATEMENT OF JANET L. YELLEN, SECRETARY OF THE TREASURY, AND SHALANDA D. YOUNG, DIRECTOR OF THE OFFICE OF MANAGEMENT AND BUDGET, ON BUDGET RESULTS FOR FISCAL YEAR 2022](https://www.whitehouse.gov/omb/briefing-room/2022/10/21/joint-statement-of-jane...

[Few-shot Learning with Retrieval Augmented Language Model](https://arxiv.org/abs/2208.03299) >Large language models have shown impressive few-shot results on a wide range of tasks. However, when knowledge is key for such results, as is the case for tasks such as question answering and fact checking, massive parameter counts to store knowledge seem to be needed. Retrieval augmented models are known to excel at knowledge intensive tasks without the need for as many parameters, but it is unclear whether they work in few-shot settings. In this work we pres...

I have to admit noticing you were an admin surprised me, but you've always been friendly and informative in my experience - congrats on the position.

@(monkyyy) If the Democrats win the House, they already won the Senate essentially. Basically, the two are correlated and the required votes to get there run through expected Senate races. More precisely, some of the House seats that are 'gettable' (or defensible) for Democrats are in these races, and more of them are in areas that are geographically/demographically similar and more pointedly they would require significantly more performance wise, and would likely move the Senate needle with them (meaning they might actually increase total Senate seats f...

@ugandamaximum Agreed, but I don't think there's a realistic chance of it resolving negatively - at this point it's basically either positive or ambiguous. Nonetheless, in this case it appears specifying a number has made this resolution more problematic rather than less.

@WilliamKiely They're generally late yes, 2020 was one of their fastest calls from recollection and still took months. Do not expect this to be resolved immediately most likely (though at this point it seems very unlikely to resolve as before 2023).

I'll be honest, I'm well off the community here. Particularly with what Lavrov has had to say about Russia's determination lately. Honestly, even the humanitarian corridor deal makes me suspect there won't be one that lasts a month before the end of this year. Good chance of an ambiguous resolution though I suspect as I continue to think they'll take Kyiv, and at least try to install a puppet government.

Forgot about this question, but bluntly I suspect Putin may have saved his career. Revising substantially upwards given the situation appears irrelevant now.

My uncertainty here is what would count as de facto control. There's a very real chance a puppet government is set up, but one that can't actually exert real authority outside militarily. Does merely driving Zelenskyy out of Kyiv and having a claimed government there count if they hold military control over the area? What if they claim legal control, but aren't able to set it up to function? If they hold legal control, but are still fighting a guerilla war over the same area? Basically, depending on what counts, I'd rate this from almost no chance whats...

This is a bit of a bizarre question, even with the workarounds the AGI question has two problems for these purposes - namely that it asks for multiple specific tests, and that the actual demanded performance is relatively low comparatively speaking. Does such an AGI necessarily mean you'd immediately get an ASI? My immediate response is no, but such an AI may be much more capable than those requirements would indicate.

https://www.nytimes.com/live/2022/02/15/world… - After meeting with Germany’s leader, Putin says some Russian troops will return to their bases.

@Jgalt Agree, particularly given this edit completely changes the question - based on the change in community forecast a lot of people were expecting a resolve on this. Honestly I might go so far as to say this should resolve ambiguous and be reopened.

@(EvanHarper) As someone who doesn't generally predict outside 5%-95%, I'm at 1%. One, there has only been one product approved to date, and simply paying off the costs of their current production facility would likely cost more than the entirety of their production capacity. Substantially more if I recall the specs and what the size of relevant funding round correctly. Two, as someone noted below they have no real incentive to even attempt this in this term - startups are not generally expected to turn a profit for at least a few years, and they just i...

@admins - Resolves as the 28th according to chart. Although I was predicting when it collapsed to 50% rather than when it capped. Oops?

I have to admit, I tend to view this question as awkward - largely due to when it was penned likely. With internal statements about preparing to take Taiwan by 2027, a breakpoint of that date makes it a poor fit as a proxy question (which is the most likely way this would resolve). I will say I'm still slightly above the community here (partially due to how easy the criteria are to fulfill), despite putting most of my weight in terms of risks of invasion of Taiwan later in the decade, but as is it largely amounts to a question (given the speed of govern...