You’ve probably heard of prediction markets in the last few years, particularly crypto-based ones. You can bet on anything from elections to sports to the year Jesus Christ will return1.
In my experience they can be fairly accurate, but not incredibly so. For instance, in September 2023 (before the current round of crypto prediction markets got big), I was trying to see if the government would shut down before the end of the year, especially since there was a major deadline of October 1st. I found a question on Metaculus that asked this, and as we got closer to the September 30th deadline the market got up to 90% (and all the way up to 95% when looking at just the pros). But then, they were able to negotiate a deal and no shutdown happened for the rest of the year. To be fair there was no money at stake (Metaculus uses reputation instead), there was still a 10% chance it wouldn’t happen, and also n=1. Still, this exemplifies how prediction markets are usually a better gauge of market sentiment than ground truth.2
My interest in prediction markets returned in early 2026 in the wake of Maduro’s imprisonment and Khamenei’s death. Previously only reputation was gained or lost, but now huge payouts were being made based on the fates of individual world leaders. All of a sudden, an old idea was back on the table and more imminent than ever: assassination markets.
Assassination markets were originally discussed in the 1990s. Jim Bell’s original proposal from the “Cypherpunks” mailing list involved bounties that could be collected by someone who submitted a cryptographically hashed document ahead of time. This document would have enough identifying information about the assassination that, upon completion of the bounty, the person who posted the bounty could send a payment to an address specified by whoever sent the document.
This system was tested in 2013 on Tor with bounties posted for Barack Obama, Ben Bernanke, and other high-profile officials but nothing came of it. The obvious failure mode for the classical setup would be law enforcement secretly making many posts with large bounties, making you commit yourself to specific details and then not only alerting the target but using that information to catch you. Even if there was some reputation mechanism, it would be hard to prevent fake transactions, sybil attacks, or other infiltration by white hats.
What betting markets do is make these markets much more liquid and impersonal. As a would-be assassin, you don’t need to rely on someone posting a large bounty in order to make a significant amount of money. You can just make a large bet that has a very good return on an outcome that resolves your way if your target dies. This includes not only resolution upon death but upon leaving office, not completing a task, or other outcomes that have “being alive” as a requirement. Thus, professional assassins could make money on targets with a high ratio of potential winnings per dollar vs difficulty of killing them and getting away with it (and the money). If the assassins fail but don’t get caught, they could still keep their options to try again another day or just sell them on the market. In fact, even a failed assassination attempt would be profitable since it could raise the market odds that someone dies, enabling the failed assassin to still sell shares for a substantial financial gain.
This opens some strange and scary questions. Could you bet against your own survival to worsen the odds any potential assassin would get, or would other rational market actors bid this back down to a more reasonable value once this becomes a known strategy? This might actually worsen things since making a bigger market may leave room for assassins to win more, or at least reduce downside risk in case they fail and sell their options. Even if it doesn’t do much to deter assassins, placing bets against your survival acts as a kind of life insurance.
Generally this becomes worse for you the more famous you are, since there’s at least a small chance that each additional person who knows about you will bet on your death one way or another (and a bigger market is more attractive for would-be assassins). It is especially bad for public officials since a lot of bets about political outcomes hinge on being alive. But strangely, how well-liked you are may play little role in determining the likelihood that you will be assassinated. Being more disliked would raise scrutiny, make many more people want to kill you, and expand the size of the market, but being more well-liked could reduce the odds of assassination and therefore make a successful assassination bet that much more profitable.
The silver lining of prediction markets is that, as a famous person, you can easily create markets for things that you can then decide which way to resolve. For example, Polymarket has bets for how many tweets Elon Musk will post in a week. As Elon, it would therefore be straightforward to place a bet that you would tweet a certain number of times and then actually tweet that number of times, making more money the more unlikely the prediction is.
Would you try to kill your enemies and get paid to do it? Even if you could somehow get away with it, this would risk backfiring through reprisals or popular sentiment turning against you. Therefore reasonable people probably wouldn’t, but once you face the prospect of being financially set for life if you kill just one person (or even just make an attempt), a tiny segment of the population is all that has to be sold on this bargain to cause a major problem. Even if you wouldn’t kill someone for a $5,000,000 reward, at least one person out there in the world probably would. See Dostoevsky’s Crime and Punishment for more on this theme.
Regardless of morality, these kinds of markets are rapidly becoming normalized. The development of truly anonymous cryptocurrency like Monero and ways to host software in a private and decentralized manner makes further steps down this road not only more likely to occur but much more resilient once they do. Short of unprecedentedly efficient global intergovernmental coordination, which could theoretically happen in the wake of an increase in assassinations or other negative effects of prediction markets, this will continue. Fortunately, this might be the one thing that unites government officials around the world, as they would all be at increased risk regardless of country or party.
Footnotes
-
my personal favorite one, because why would you be trying to earn money if you really thought Jesus was coming back? ↩
-
Also you may not consider it super important but it was for me since I was working for the government at the time, and I was also planning on visiting a national park which would have been closed if the shutdown went through. ↩