Basic game theory and how you can do a bunch of good in ~3 Hours. (developing article.)

Epidemic status: this is a bit rushed at the moment, and It’s missing some stuff, but soonish, it‘ll be entirely correct, unless I make a bunch of human errors repeatedly, missing an error I made after plenty of review. PLEASE leave comments as to how I can improve this article instead of downvoting it. (Update: the editing process is delayed to make room for other stuff in my calendar)

everyone makes decisions on these three things:

their information,
their decision-making process (which you would hope would be “whatever best fits their values”, but we have to account for things like conflicts of interest/human error and whatnot.)
what options they have.

out of all [the options] they have, they choose based on their [decision-making process] which outcome they think would be best, based on [the information] they have.

If you want to influence what someone does, you HAVE to influence at least one of those things.

#1: while it’s obvious that people don’t act under knowing everything, this is of huge importance, and is why spies exist. if you influence what a world leader thinks is true, it has the same effect on their decisions in a scenario where the world is like what you told them.

It’s also very easy to give important people information: many of them have public e-mails, phone numbers, staff, etc. Also, they probably work at a place, such as the US capitol, the UN buildings, Moscow’s capitol equivalent, etc, and they have to enter and exit said building. The UN has a public dining space with many representatives that go there daily. Maybe you can strike up a conversation with them? (https://www.un.org/en/desa/contact-us) You can also leave a sign outside the buildings, or maybe holler a URL at them as they enter and exit said building. (That last one might not be great, since many of them have bodyguards, and it also is nerve-wracking to try.)

for #1, you can sort of force them to get information, say, by giving them a speech or shouting at them or whatever else. Also note that, from the perspective of a person with some goals, they pretty much always want more information, since they can usually pretend they didn’t hear it if they think it would be bad for them. (except for the edge-cases where you can’t ignore it, even if you think it’s negative to use for your values. This happens when your decision to ignore it is impaired by non-logical factors. Someone with a drug addiction might try to block out information about how to get drugs.)

However, it might be bad for them that now you know that [they now know what you told them]. For example, if you tell someone who just WONT. STOP. ARGUING WITH YOU. about the sunk cost fallacy. now, whenever they use the sunk cost fallacy, you can point out to them how they made a mistake and they did it while knowing that they shouldn’t.)

While you could tell someone true information, you can also tell them misinformation or disinformation, and you can always make something that’s disinformation or misinformation to become true. That’s what a strike is. It makes something not true (that a union is on strike) into something true, in an attempt to change what they choose to do. Lying has some moral objections in some ethical codes, however I would argue that consequentialism largely makes lying allowed when it comes to important people, and there’s plenty of arguments for consequentialism. Lying especially makes sense when it stops even more lying, such as telling Narendra Modi (current prime minister of India) that the US has discussed plans to give India more funding if they manipulate the media less.

Another clear example when lying probably makes sense is when the stakes are nuclear war.

“Information” mostly refers to physical features that aren’t set in stone. Things that are set in stone (like math and philosophy) are what I would say better falls into the category of #2. That includes probability things. For example, “there’s a 50% chance I will give you three gummies if you give me one gummy” counts as “information”.

Effective Altruism does plenty of stuff with the UN all the time, and if you want to directly say something but you’re really far away, you can ask an Effective Altruist to do it for you, whether it be the UN, the EU, the capitol of the US, etc.

#2: Influencing someone’s decision-making process is much harder, but it’s still basically just communication. you mostly just try to convince them of some decision-making method. do note that they would decide to either change their decision-making process to some of what you recommend or not to change their system USING their CURRENT system; you wouldn’t want to take a pill that makes you want to kill your kids, and someone who really really really has one opinion that they hold strongly (e.g., who to vote for) might not be willing to change their mind.

It’s MUCH HARDER to force someone to change their decision-making process. You might think that testing them by putting them through imaginary or real scenarios and see what they choose to do, but remember: They’re probably taking into account that you might be doing this, and that might make them change their decisions to something that they want you to think they would do. namely, they’d want to pretend they’ve changed to agree with you.

They might also take into account that you’re talking THAT (they take into account that (you take into account what they do)) into account, and you should account for that. they might account for THAT THAT as well, etc.

It’s usually much easier to get them to voluntarily change their decision-making process, especially to one that gets rid of illogical decisions, but doesn’t change their values (since you’d have to convince them to change their values WHILE THEY HAVE their current values, and a world where they have different values doesn’t match their values as much as a world where they don’t): tips on how to stop procrastinating, some common logical fallacies, how to make better decisions using meetings, etc.

Many people value making rational decisions, so if you provide a rational reason as to why your decision-making method is more logical than theirs, then you can change most of their values by appealing to their need to be logical.

A common pitfall here is when someone’s value system rejects complicated logical arguments. Someone who is a “no-nonsense traditionalist” might not be willing to listen to far-fetched arguments, labeling them as nonsense. Here, one strategy might be to point out a logical inconsistency in their decision-making system, as that one is often seen as not far-fetched.

For example, the belief that the life of a person you know is more important than the life of someone else (commonly phrased as “you should do good in YOUR community” is incompatible with the belief that everyone is equal.

A good way to scale this up is to get the UN to get all the world leaders to agree on a moral code (https://forum.effectivealtruism.org/posts/FtznFEKvhYEM9EChF/it-s-time-world-leaders-get-their-values-aligned-here-s-how) or otherwise agree on a decision-making system for them all to follow. In the cold war, either the US was right to go against the USSR, or the USSR was right to exist, or neither, but probably not both. If they just agreed to who was right sooner, that probably would’ve lead to it ending sooner. One could reasonably argue that Gorbachev thought the US was right.

Do note that world leaders and important people are very busy, so if you send them any info or reasoning to change their decision-making system should be fast, and should be enjoyable enough to consume that they’ll consume it. If you have something very important to say, it might be worth it to pay Rational Animation’s animation team to make a video of it, or use CapCut, etc.

#3: Changing someone’s options is much, much harder.

Here’s some examples:

This is what an election is: it determines who gets to have the options of someone in office.
It’s also mostly why people kill other people (except for the much more common cause, which is illogical decisions, such as being crazy, being in the heat of the moment, senseless pride towards being in the military, etc; and the sometimes logical, but usually not logical decision to commit suicide.)
It’s also largely what wars are about.
It’s why people get money: it gives them the option to use it. There are edge-cases where someone having money effects other things, such as how much paper there is, or (if the person is Mr. Crabs) if the person inherently values money, etc.
Other
On elections: there’s plenty of ways to effect this, and there’s plenty of places to learn how to do so. Here’s a handful of stuff:

this message from the Effective Altruism slack channel:

“Here are some volunteer opportunities to help with the election which I think are unusually impactful. If you’re wondering whether election work is worth spending time on, here’s https://www.lesswrong.com/posts/GXSMD84DgjJ7ch9T4/why-the-2024-election-matters-the-ai-risk-case-for-harris for why working on the election is likely among the most impactful options at the moment (here’s the Ehttps://forum.effectivealtruism.org/posts/QJfy2gBdZbw5cDXdh/the-case-for-contributing-to-the-2024-us-election-with-your#4pmpXWPzLSgouKEJq, though it’s a bit outdated).

One-off volunteer opportunities https://www.mobilize.us/ml-volunteer-hq/event/696232/: The work involves connecting to an autodialer which will connect you with people who have partially completed their voter registration documents but have not yet fully registered. These calls are simply reminders to complete the registration and to get out and vote. Compared to normal phone-banking this is an unusually friendly audience and folks are often glad to be reminded to finish registering to vote.

https://www.mobilize.us/ml-volunteer-hq/ For this project volunteers go to local bars and offer free beer to patrons for texting 5 friends living in swing states. This is a fun way to spend an evening and we think it’s highly impactful.

Benefits from weekly engagement https://linktr.ee/studentturnoutprojects is an organization that supports and coordinates four swing state campus GOTV groups.They’re currently looking for virtual volunteers to support the overworked campus organizers with critical behind-the-scenes tasks. This helps free the organizers to do more crucial in-person work.

To get started you can attend one of the Tuesday or Thursday evening calls to get introduced. https://lu.ma/stp If you can’t make one of those, set up a chat with the team lead (Arden) https://savvycal.com/arden/election?return_to=https://savvycal.com/arden&d=30&view=week&from=2024-09-15&sid=6e3d5ace-be44-415a-9661-f38b8afd750f.

It’s helpful to be able to join on a regular basis (basically for the next 4 weeks) but one off is fine too”

https://youtu.be/zE7GsHoBedc,

https://youtu.be/rStL7niR7gs. There’s definitely more on this that I didn’t include.

while number 2 is one way of doing this, it’s highly morally objectable under most moral codes, and very risky, and could certainly lead to a bad, unintended outcome, such as a massively tarnished reputation on Effective Altruism, or the classical example of how the murder of Franz Ferdinand led to WW1, which led to WW2, which led to the cold war and much of the tension in the middle-east, which leaves us here. Also, this article is NOT saying you should kill someone. It is merely recognizing why some might consider it. I do NOT endorse killing, with a few edge-cases, such as when it’s a terrorist and killing them saves lives. again, NOT endorsing murder.
It’s pretty tricky to effect war, and if you can, good for you! otherwise, you might try to effect war by using [parts of how people make decisions] #1 and #2.
There’s plenty of ways to get money, and there’s plenty of research on that. You can also bribe some important people, but that’s often morally objectable, and is risky-illegal in some places. In some contexts, it’s a friendly 20 dollars that says you can’t eat a philly cheesesteak in under a minute, but the stakes are much lower in those contexts.

This applies not only to people, but to anything that causes something. Here are some slightly abstract examples:

An AI: an AI’s information is the inputs it has, and AI’s decision-making process is its weights and functions and code and whatnot, and its options are the set of all of its possible outputs.
A country: its information is just the information all the people who directly effect (e.g., work at) the country’s actions, its decision-making process varies from country to country, but in, say, the US, it’s primarily how a bill becomes law, how the court system works, how the budget is decided, and what the executive branch can do. It varies from country to country what their options are.
Dice: a dice is “informed” by the physical interactions around it (e.g., wind). Its decision-making process is mostly the physics of how dice rolls, and its options are mostly to either land on a 1, 2, 3, 4, 5, or 6.

Sorry for the grammatical errors.