Update?
Bojadła
Karma: 102
Invariably, within 10 minutes, I receive messages from coworkers congratulating me on my boldness. You asked him? That question? So brave!
How do you respond to those messages?
Could you make it more clear how you would change the original Harris market to avoid the problem? I’m not sure what exactly you mean by “base question and a conjunctive market”. I figure you’re talking about the “other” option in the market you link, which prevents N/A. But how do you add that to the Harris market? Is it just adding one more option “Another person not listed here is nominee and Harris wins”?
I read your post but didn’t understand it immediately. After thinking about it, I wrote what I feel the main points are in my own words. Maybe this makes it more clear to others too. Or maybe the OP will comment that I misunderstood something.
The “Harris market” can be seen as (is functionally equivalent to) a collection of markets. Each market has the following form:
Would Harris win, if {person} were her Vice Presidential nominee in 2024?
Each market resolves as follows:
Yes, if {person} is the nominee and Harris wins.
No, if {person} is the nominee and Harris loses.
N/A, if {person} not the nominee.
Each market is independent. I mean this in the sense that there is no manifold mechanic that ties them together, except that they are visually grouped under “the Harris market”.
There is a problem with this kind of market. When people see a high probability of a person in the market, they mistake it as meaning that the person is a likely choice or a good choice.
This is a mistake because it does not follow from the market’s incentives. Imagine someone that will obviously not be nominee, like “Batman”. You know that Batman will not be nominee, so you know that this market will resolve to N/A. This makes bets on Batman meaningless. You can bet Yes or No, it doesn’t matter. The market will resolve to N/A and you get your money back. Some people do this because they find it fun to see that Batman has a high Yes probability.
Batman can’t be nominee because he doesn’t exist, but you have the same problem for obviously terrible candidates. Imagine the person most hated by all potential Harris voters, “the Joker”. (Assume Joker was a real person.) The true probability of Harris winning with Joker is tiny. Yet, in the market, there is little incentive to correctly predict this. Everyone knows Joker is a terrible choice, so Harris is not going to choose him as nominee, so the market will resolve N/A.
The author of the Substack you quote makes this mistake. They did not consider what the market incentivizes.
They also made a separate mistake that applies to all prediction markets, not just misleading conditional markets. They did not take into account how much money it takes to intentionally influence the market. It is easy (cheap) for people to push small markets (markets with little money invested) in their desired direction.
I agree and I am putting my money where my mouth is.
I will play this game under the rules linked in the OP with me as the gatekeeper and anyone as the AI. I will bet at odds 10:1 (favorable to you) that I will not let the AI out. The minimum bet is my 10 k USD against your 1 k USD and the maximum bet my 100 k USD against your 10 k USD. We will use a mutually accepted LW community member as referee.
If you believe you have at least a 10% chance of making me let the AI out, you should take this bet. I predict no one will take me up on it.
I speculate that the reason that gatekeepers let the AI out is that they do not take the experiment seriously enough. They don’t really care about not letting the AI out. Maybe they want the outcome to be that the AI has been let out so that people take AI safety more seriously. I’m not saying an actual superintelligence couldn’t do it, but no currently available intelligence can (with me as the gatekeeper).
The bet has been lost. The merge wasn’t allowed. https://archive.ph/MDYWF
The share price of Spirit Airlines (NYSE: save) dropped to 50%.
I consider it unlikely that my interactions with ChatGPT 3.5 (that led to my original post) played a significant role as training data that helped make ChatGPT 4 so much better.
Why do you consider it unlikely?
If I was good at memorization (model parameter size) but bad at reasoning then your original post showing up in my training data would help me.
I’m also curious how much of the improvement comes from improvements to the hidden and secret prompt that Openai adds to your Chatgpt interactions. Unfortunately we can’t test that.
Also, to be clear, nothing in this post constitutes investment advice or legal advice.
I often see this phrase in online posts related to investment, legal, medical advice. Why is it there? These posts obviously contain investment/legal/medical advice. Why are they claiming they don’t?
I guess that the answer is related to some technical meaning of the word “advice”, which is different from its normal language meaning. I guess there is some law that forbids you from giving “advice”. I would like to know more details.
Edit: This question was answered in a previous comment.
You are implying that it is hard to get Samsung expose. Why? On their website [1] they list several ISINs. Some of them I can buy in through my usual broker. They aren’t special.
[1] https://www.samsung.com/global/ir/stock-information/listing-Info/
My problem is that you link libsyn in the post but it doesn’t contain all episodes. I was able to find the other episodes but I would still like for the post to be updated.
Are you no longer updating the libsyn version? The most recent episode there is from August 2022 while the other sites have newer episodes.
Is it possible to buy the book as an ebook? Why not?
You can find the previous volumes online at https://www.lesswrong.com/bestoflesswrong .
Do you OP have access to secret (non public) information related to the bet?
The link to the 36 questions is paywalled. I am pasting them here.
Given the choice of anyone in the world, whom would you want as a dinner guest?
Would you like to be famous? In what way?
Before making a telephone call, do you ever rehearse what you are going to say? Why?
What would constitute a “perfect” day for you?
When did you last sing to yourself? To someone else?
If you were able to live to the age of 90 and retain either the mind or body of a 30-year-old for the last 60 years of your life, which would you want?
Do you have a secret hunch about how you will die?
Name three things you and your partner appear to have in common.
For what in your life do you feel most grateful?
If you could change anything about the way you were raised, what would it be?
Take four minutes and tell your partner your life story in as much detail as possible.
If you could wake up tomorrow having gained any one quality or ability, what would it be?
If a crystal ball could tell you the truth about yourself, your life, the future or anything else, what would you want to know?
Is there something that you’ve dreamed of doing for a long time? Why haven’t you done it?
What is the greatest accomplishment of your life?
What do you value most in a friendship?
What is your most treasured memory?
What is your most terrible memory?
If you knew that in one year you would die suddenly, would you change anything about the way you are now living? Why?
What does friendship mean to you?
What roles do love and affection play in your life?
Alternate sharing something you consider a positive characteristic of your partner. Share a total of five items.
How close and warm is your family? Do you feel your childhood was happier than most other people’s?
How do you feel about your relationship with your mother?
Make three true “we” statements each. For instance, “We are both in this room feeling … “
Complete this sentence: “I wish I had someone with whom I could share … “
If you were going to become a close friend with your partner, please share what would be important for him or her to know.
Tell your partner what you like about them; be very honest this time, saying things that you might not say to someone you’ve just met.
Share with your partner an embarrassing moment in your life.
When did you last cry in front of another person? By yourself?
Tell your partner something that you like about them already.
What, if anything, is too serious to be joked about?
If you were to die this evening with no opportunity to communicate with anyone, what would you most regret not having told someone? Why haven’t you told them yet?
Your house, containing everything you own, catches fire. After saving your loved ones and pets, you have time to safely make a final dash to save any one item. What would it be? Why?
Of all the people in your family, whose death would you find most disturbing? Why?
Share a personal problem and ask your partner’s advice on how he or she might handle it. Also, ask your partner to reflect back to you how you seem to be feeling about the problem you have chosen.
Could you recommend (or link to a recommendation) of what first aid kit to buy? What should it contain? How expensive should it be? Where should I buy it from?
It is 2022-10-16, roughly one year after the post was made. The S&P 500 has not dropped below 3029 at any point. It is currently at its lowest value since the post was made at 3583.
https://www.spglobal.com/spdji/en/indices/equity/sp-500/#overview
Which resources do you recommend to start learning meditation this way?
This might be “optimal given your state of knowledge about ROB” but it isn’t “guaranteed to succeed more than 1⁄2 of the time”. For example:
My prior is that Rob always picks a positive and a negative number.
Actually Rob always picks two positive numbers.
Thus when we play the game I always observe a positive number, guess that it is the larger number and have a 0.5 probability of winning. 0.5 is not greater than 0.5
You asked people to try resolving the problem on their own. So here is my answer before I read further. I was motivated to find a reason for the doomsday math to be incorrect by your prompt.
I felt the mistake was the hidden ambiguity of this statement:
How do we randomly sample the person? Imagine you had to code this as a simulation in Python. Which of the following algorithms does your program follow?
Pick uniformly randomly between Short and Long. If Short, then pick a number between 1 and 100 billion. If Long, then pick a number between 1 and 100 trillion.
Create a list [“Short-1”, “Short-2“, …, “Short-100-billion”, “Long-1”, “Long-2”, …, “Long-100-trillion”]. Pick a uniform random sample from this list.
The math you present matches the first algorithm. The math is different with the second algorithm. With the second algorithm we have:
P(sample beginning with Short) = 1/1001
P(sample beginning with Long) = 1000/1001
P(sample ending with 6 | sample beginning with Short) = 1⁄10
P(sample ending with 6 | sample beginning with Long) = 1/10000
P(sample ending with 6) = P(sample ending with 6 | sample beginning with Short) * P(sample beginning with Short) + P(sample ending with 6 | sample beginning with Long) * P(sample beginning with Long) ≈ 0.0002
P(sample beginning with Short | sample ending with 6) = P(sample ending with 6 | sample beginning with Short) * P(sample beginning with Short) / P(sample ending with 6) = 0.5
This matches the following experiment:
Run the program to pick a random sample.
If your sample does not end with 6, then restart.
You have a sample ending with 6. Check whether it begins with Short or Long.
You observe Short and Long with equal probability of 0.5 .