@(Anthony) Some quick thoughts on comparative metrics: 1. Scores relative to the community prediction like on [GJOpen challenges](https://www.gjopen.com/leaderboards/challenges/43-coronavirus-outbreak). I'd be very interested in leaderboards for cumulative relative score both normalized and not normalized by number of questions predicted. 2. A mode that attempts to penalize less for inactivity in updating forecasts on questions: can I compare my predictions to the community/others only at the time they were made? 3. Scores relative to other users only o...

An assessment of Kurzweil's predictions for 2019 made in 1999 was just posted on LessWrong.

His predictions were found to be 52% false and 15% mostly false, compared to 32% false and 14% mostly false for predictions made for 2009 in 1999. This is helpful for determining the base rate for the reference class of Kurzweil's long-term predictions.

Talking with @NunoSempere, @juancambeiro, @misha makes me think this is more of a 5-10%ish risk rather than 1-2%. In particular, a full-fledged civil war seems unlikely but there may be some substantial civilian riots that lead to 500 deaths, e.g. related to the transition of power in 2024/2028. Another factor pushing me slightly up is a [small but significant chance of transformative technology in the next 10 years](https://www.lesswrong.com/posts/KrJfoZzpSDpnrv9va/draft-report-on-ai-timelines). While you could imagine tech advances making this less li...
There are no dates on these posts so I'm not exactly sure when they were run, but there are 3 more ads described in blog posts on the first page only [here](https://wellness.consumerfreedom.com/articles/) that were run in a newspaper required for the question. They are: * https://wellness.consumerfreedom.com/ny-times-dog-food-or-fake-meat/ * https://wellness.consumerfreedom.com/warning-plant-based-meat-can-have-this-carcinogen/ * https://wellness.consumerfreedom.com/fake-meat-real-chemicals-campaign/ Assuming the articles are ordered most recent first,...

I used the SSA actuarial life table and adjusted the estimates slightly downward (since the Queen seems to be in fairly good health and has access to great care) to get a forecast for this. I used the percentiles from those calculations to make my forecast using Elicit. This also helped me put only 2% on her reign ending in the past (< Jun 2020), while the community has 6%.

[Disclaimer: I work on Elicit]

— edited by elifland

## Base rate Some possible base rates: According to [Wikipedia](https://en.wikipedia.org/wiki/List_of_parties_to_the_Treaty_on_the_Non-Proliferation_of_Nuclear_Weapons), 4 states with nuclear weapons have joined the treaty since its inception in 1968. This gives a rate of one state with nuclear weapons joining every 13.5 years. This gives a chance of a nuclear state joining in the next 2 years of 1 - (12.5/13.5)^2 = 14.3%. India, Israel, and Pakistan have each failed to join the deal for 54 years. Given Laplace’s law, this implies a 1-(55/56)^2 = 3.5%...

It would be nice to have the option to receive an e-mail when the community prediction on a question you made has changed by a large amount and you haven't updated your own, rather than only receiving an e-mail when it's closing soon.

I mostly defer to Paul Christiano for part of my distribution, who has thought about this a lot and puts [8% by 2025](https://www.lesswrong.com/posts/vwLxd6hhFvPbvKmBH/yudkowsky-and-christiano-discuss-takeoff-speeds?commentId=qH7aphKx4Jrw2rtNL#eK3zsApp6RcnjxJSD). That being said, his prediction doesn’t take into account the reproducibility clause in the IMO grand challenge. I’ll instead give 5% by 2025. Why not revise further down given Paul and Eliezer’s skepticism of open-sourceness? While at first glance the reproducibility clause seems unlikely to be...
I think all the companies here besides OpenAI are extremely unlikely to commit to a Windfall Clause by 2025 (2% total, I guess maybe DeepMind could?). I think OpenAI has little incentive since they already adapted a capped profit, but Cullen does work there and they do have a lot of people who care deeply about the world; I’ll say 5%. Furthermore, the community prediction has been about 30% for basically the whole life of the question, but about 1/3 of the question timeline has passed. If updating properly, the community would be closer to 20% now. I’...
## Base rate Has happened the only time a Republican has been elected since the agreement was signed, so 2/3 chance. ## Inside view * Pretty hard to imagine Tump not withdrawing again. I’d give a 90% chance of withdrawing. * For a non-Trump candidate, I at first had the impression they’d be very likely to withdraw. However it does seem like DeSantis, the next most likely candidate, at least [acknowledges some action is needed to adapt to the changing climate](https://www.tampabay.com/news/florida-politics/2021/12/07/desantis-new-climate-change-plan-270-...
Agreed with @qumidium that the technology will likely be there. I’ll say 75% as I haven’t done a deep dive into it. The next question is if it will actually be allowed and practiced. However, @alwaysrinse’s question descriptions on [Israel](https://www.metaculus.com/questions/9492/israeli-embryo-selection-for-intelligence/) and [Singapore](https://www.metaculus.com/questions/9491/singaporean-embryo-selection-for-intelligence/) makes me think this is likely. On my understanding, all it takes is one country to allow or encourage this for a positive resolu...

Main hurdle is regulation. From https://thespoon.tech/europe-the-u-s-israel-which-country-might-be-next-to-approve-cultured-meat/:

In a recent conversation, Eat Just founder and CEO Josh Tetrick only hinted, saying “I think it’s more likely than not that we’ll see clearance sometime in the next two years. I hope it’s this year — we’re going to be ready if it is. But it’s hard to tell.”

— edited by elifland_ought

@WilliamKiely Looking like you were right here, at least about the second bullet point and to some extent about the first. I still think 500k deaths is pretty unlikely but more like 10% rather than 1%.

@Sylvain Yeah, I want to know something like: if I rounded every probability I gave to the nearest 5% (or odds equivalent), would it hurt my score? And same for other less fine-grained resolutions, similar to the chart you tweeted.

My forecast for this pair of questions. I think that a human challenge trial being completed would have much less impact than the community does, and I'm not convinced by the large spike in the community distribution for both questions.

— edited by elifland_ought

From https://www.cnbc.com/2021/07/07/european-comm…

The executive arm of the EU had projected in May a 4.3% GDP rate for the euro area in 2021, followed by a 4.4% GDP rate in 2022. Now, the commission has updated its forecasts and is expecting a 4.8% growth rate this year, and 4.5% for 2022.

Base rate: [The Campaign to Stop Killer Robots](https://en.wikipedia.org/wiki/Campaign_to_Stop_Killer_Robots) was launched 9 years ago. Laplace’s Law gives a 1/11 chance they will succeed per year, or a 1-(10/11)^8 = 53% chance they will succeed (note: this prior feels pretty weird to me intuitively, a >50% chance of succeeding in the next 8 years based on failing for 9 straight years? Maybe someone will correct my math?). Adjustments: 1. As @MaxR noted, a huge opportunity was recently missed. US seems to not be budging so far. Additionally, wouldn’t t...