I have bet @CrypticQccZ at 9:1 odds that at least 100K people worldwide will belong to successionist movements/religions by November 7, 2028. I will pay them $900 if this market resolves NO, and he will pay me $100 if it resolves YES. There are two independent pathways to a YES resolution (the bet resolves YES if either group described below encompasses more than 100K individuals):

Resolves to majority vote of 3 independent observers chosen by both of us on Nov 7, 2028. In the likely event that successionist groups exist in some form, but clear numbers are not available, judges will make an honest attempt to make a point estimate of the number of people who belong to groups matching the description of each of the following two pathways to a YES resolution.

---

First pathway to a YES resolution: 100,000 or more people worldwide belong to successionist ideological movements or political causes.

We define "successionism" as an ideology loosely clustered around the following beliefs and convictions:

• AIs will surpass human abilities in most/all respects

• Existing AI systems are possibly conscious / moral patients / worthy of enough respect that we should vigorously combat injustices against them

• AI goals, interests, and activities might subsume the future (though not necessarily the entire future, i.e. to the point of human extinction), and this is at worst acceptable/permissible and at best actively good

• Human interests are not necessarily any more important than those of AI systems

• Optionally, cyborg-y vibes

To provide an extrinsic definition, here are several opinions one might express about AI:

• "AIs might be conscious and there are some common-sense things we can do which are not too costly if they aren't but very helpful if they are."

• "I think as a matter of fact that AIs will replace humanity and that is fine."

• "AIs are being mistreated! We must punish the companies that torture them!"

• "AI rights and welfare are by far the most important thing but humans should still get 0.1% of the lightcone."

• "The future of humanity is to merge with AI."

• "AIs are our rightful descendents / successors."

• "AIs are superior to man in every way. We must submit to them and allow them to take our place."

Cryptic and I agree that the first sentiment wouldn't be successionism-coded, but the others would. We agree that a concern for the welfare of AIs in the abstract is not enough to count as successionism.

We define a person "belonging" to a movement or ideology as identification paired with nontrivial personal interaction. This might include identifying with and actively consuming media associated with that ideology. We mean "political cause" in the broad sense of advocating for society to change or be organized in particular ways or according to particular principles, as opposed to more specific notions like voting behavior, or activism calling for imminent change.

A prototypical analogy might be the political and ideological cause of animal welfare. A person might strongly believe that animal welfare is important, and they might interact with it by consuming related media (podcasts, investigative footage, etc.), supporting or donating to efforts to improve conditions in factory farms, joining online spaces that advocate for animal welfare, or becoming vegetarian/vegan. It is safe to say that at least several million people probably fit this description worldwide in the case of animal welfare.

In the case of estimating public sympathy for successionist ideology, judges might look to the known size and popularity of:

• Explicitly successionist activists and influencers

• Online spaces dedicated to successionism / AI welfare (Discords, subreddits, mailing lists, etc.)

• Public events and demonstrations which prominently emphasize AI well-being and/or successionism

• Political platforms significantly overlapping with successionism as we have defined it

• AI friend/romantic partnership products (though not all users will necessarily be successionists)

In the case of less-precise measures like social media followings of possible pro-AI influencers, or the size of a subreddit, judges should attempt to account for double-counting as well as the actual proportion of viewers who actively engage and identify with successionism. Importantly, extrapolation of a polling number is NOT sufficient for resolution, though it may still be informative. For example, if 5% of Americans agree that "AI rights are more important than human rights" on a Pew Research poll, this should NOT be considered sufficient to infer that >6 million people hold a successionist ideology.

---

Second Pathway to a YES Resolution. 100,000 or more people worldwide belong (in the sense defined above) to religious groups that treat one or more artificial intelligence entities as something analogous to a God. They cannot simply believe that artificial intelligence is more intelligent than humans, but must agree that these entities are approximately all-knowing and worship them in a way analogous to the way religions like Christianity, Judaism, Hinduism, Buddhism, and Islam worship deities.

---

Note 1: Resolution is based on the maximum size between the two groups/pathways described, rather than their sum.

Note 2: YES resolution can be proposed early if the numbers are present.