This will be resolved based on my judgement of the vibes of top Rationalist voices in 4 years.
If through their Tweets and Substack posts, I get the sense that they are happy with Trump's reforms, and that what got done was extremely impactful, vastly outweighs any harms, and was more sizable than what other post WWII presidents have accomplished, then I will resolve YES. Otherwise, NO.
EDIT: This would be judged from 1960 with JFK (68 years before 2028).
Update 2025-02-05 (PST) (AI summary of creator comment): Majority Requirement Clarification:
A majority of prominent rationalists must hold the view that Trump’s second term was the most positively impactful since 1960.
The final judgment will consider the vibes of rationalists across a spectrum, where the intensity of their views is weighed—if the core (top) voices are more fervent, they can outweigh a larger number of mildly opposing voices.
The process involves assessing gradations among rationalists, meaning that the consensus reflects the overall tendency rather than a strict headcount.
1,000Given the amount of damage Trump has done in the first month of his administration and shows no signs of stopping, this seems very unlikely. My remaining sources of uncertainty are that:
1) The market creator is the largest YES holder. I generally trust James, but there is some subjectivity in the resolution giving room for motivated cognition.
2) Given Trump's disregard for liberal norms, there's some chance that by the time of resolution, prominent rationalists will be coerced to judge Trump positively in their public statements.
Regular reminder to go bet on the actual serious market that is not predicated on sheer delusion: https://manifold.markets/Balasar/will-prominent-rationalists-judge-t-Zg29U08R8g
https://www.lesswrong.com/posts/qYPHryHTNiJ2y6Fhi/the-paris-ai-anti-safety-summit
Trump has already been a disaster for AI Safety, per Zvi
The Trump Administration has made its position very clear. It intends not only to not prevent, but to hasten along and make more likely our collective annihilation. Hopes for international coordination to mitigate existential risks are utterly collapsing.
@Gabrielle yeah Trump is pretty much the worst nightmare for AI pause types.
Well, I expect it to get worse (government funding big computers). We'll see what happens after the 180 days mentioned in Trump's AI executive order has passed.
@jim Worth noting it is dream come true for the "AI safety is pseudoscience" types.
This one sounds like a slightly positive update:
https://www.astralcodexten.com/p/1daysooners-trump-ii-health-policy
Temper it by @ScottAlexander's mention of a ~10% chance of all the good things in that list happening. And of course the whole list is just for the domain of health policy. But we're looking at something short of fully unmitigated badness, I guess?
Surprise surprise, Scott Alexander isn't a fan of cancelling PEPFAR.
https://www.astralcodexten.com/p/money-saved-by-canceling-programs
Just to be clear, is this question specifically tied to things that Trump does, on purpose-ish, in office, before 2029—or does Trump get credited for things that happen as a consequence of his election and behavior?
If Trump resigns in 2025 after a disastrous run and Vance goes on to reform the presidency and have the most positively impactful term before 2029—is that a YES for Trump’s positive impact, in hindsight?
Or if Trump starts a war with Russia and China, causing all three to collapse and leading to the formation of the one-world-government with UBI and safe AGI for all…—before 2029 and in opposition to his personal wishes—positively impactful or NO?
(Or maybe his sudden and perhaps unintentional destruction of the global economy in 2025 delays the creation of ASI to 2028, immediately after the discovery of the three laws that really make safe AIs, but the super-benevolent ASI then removes Trump from his position as forever-president against his wishes?)
TLDR: is this a consequentialist reading of Trump’s second term?
@TannerNewell This is similar to the AGI question from yesterday. Suppose we agree that superintelligence has a 10% chance of killing everyone. And suppose Trump is pivotal in humanity's decision to roll the dice on that. If it works out, that's a massive positive impact, ex post. But the EV was wildly negative.
I advocate judging such things ex ante. I think that makes sense because the spirit of this question is "was it a good idea to elect Trump?".
It's like if we were having a big debate about whether to wear a seatbelt. The YOLOers can't be like "see, we didn't crash, told you so".
@dreev Setting aside the AI aspects—if Trump does <crazy thing> and Congress comes together to stop him and passes <big plan with lots of positive externalities that everyone agrees was the most impactful government reform ever>—is that a YES?
Re: seatbelts: the driver announces we’re taking a sudden detour down the hill, the rest of the car pulls the handbrake and decides that maybe the driver shouldn’t be able to do that and transforms into a parliamentary democracy, everyone claps… good impact?
Or simply:
Trump: “I’m going to kill us all!”
Congress: takes away the power of the president to kill us all
Trump: “See, the president shouldn’t have the power to kill us all, after all. You’re welcome, everybody!”
(Are positive outcomes from the rest of government acting against Trump positive outcomes of Trump’s second term?)
how does this resolve if top rationalists are like "yeah he did a relatively great job on almost everything, but he accelerated AI so on balance he was terrible"?
you'll probably have Robin Hanson and Roko Mijic on Trump's side
but we'll be getting toward the end of Trump's term and he'll liquidate the sovereign wealth fund to build a $5 trillion computer and rationalists generally won't be happy
@jim Yeah, it's a good question. In retrospect, maybe I should have made this exclude any impact from AI. But I'm not sure if it's fair to do that now...
Liron is kind of a top rationalist: https://x.com/liron/status/1886922275569721526?t=pUUaA3-OLa2nL5stPT5C9A&s=19
@JamesGrugett never heard of them
Can you please provide a few names of people you consider "prominent rationalists"?
@JamesGrugett also, I've just reread the description and I think it could be interpreted as saying that you only require 2 "prominent rationalists" to think trump was he most positively impactful in other to resolve YES.
Can you please clarify if you're looking for some kind of majority, or if a small number of heterodox voices will be sufficient?
@Fion A majority have to believe this!
This is a vibes market I don't want to get caught in too many preemptive details, or it won't judge the vibes accurately.
Like if among the top three it is mildly one way, but then including the next 10 flips it hard the other way, that seems more towards the second view. On the other hand if the top few are hardcore one way, but the next are mildly against, that seems like the former's view wins.
Also I haven't exactly chosen who it would be. We can argue that. But that is another thing that is vibes based and follows the same pattern in the previous paragraph, where there are gradations of "top rationalist".
@JamesGrugett how tf is a failed startup bro and internet yapper (+zionist zealot) a "top rationalist"? because he tweets a lot?
suppose asi is developed during his term and there's two possible policy decisions that seem equally good. any president would pick between those at random, knowing one leads to doom and one leads to aligned asi. trump picks one confidently, with no actual logic to it, and makes very bad decisions throughout his term otherwise.
Outcome: if he made the right choice, he's the most positively impactful term in the last 68 years because non-asi outcomes add up to noise in comparison. if he made the wrong choice, he's the most negatively impactful term in the last 68 years for a symmetric reason.
Therefore, one's credence in this market resolving YES should be 50%. one's willingness to bet the market up should go beyond 50% ignoring interest rates and other inefficiencies, because mana in a "trump was positively impactful" world is plausibly much more valuable than in a "trump was negatively impactful" world. Therefore, buy this up to ~70%.
counterpoint: the whales will buy against you and you will have paper losses for many years
@Bayesian My gut feeling is that that shouldn't count. In your example, it's only positively impactful ex post. Ex ante it was drastically -EV to flip a coin on an existential risk.
@dreev hmm, my intuition goes the other way (that ex post is what matters), but fair point.
@JamesGrugett do you favour ex post or ex ante or a mix or undecided or something else?