Re: the airplane proposal, I would worry that people are too panicked in an evacuation to think through the scenario rationally, and would still grab their stuff out of instinct, possibly even slowing down evacuation by trying to then re-stow it. It’s also not a repeat game in that a single person is very unlikely to evacuate an airplane multiple times, so people wouldn’t be able to learn from their mistake (unlike the examples in your post with children). It’s a great idea for rational actors, but I don’t think people are rational actors in an evacuation. It still might be worth experimenting with the idea though to see if it works.
Lee Bousfield
Karma: 12
there is no clearer or more convincing evidence [of fair market value] than being unable to sell assets to cover even 5% of the wealth you supposedly have
The rest of your comment stands, but liquidity can be very different than valuation. To give a couple examples:
Private equity and private credit funds have restrictions on how much money can be redeemed per unit time, and can even temporarily suspend the ability to redeem from the fund if there’s underlying liquidity issues. The inability to redeem does not mean the fair market value is $0, even if there’s $0 of liquidity at the moment.
Startups often have clauses preventing anyone from selling equity without approval from the board. Investors might still be investing at $x/share but share holders can’t realize that value.
I’m pretty sure it’s literally impossible for the crash criteria of this bet to be met at this point. Based on reported information, OpenAI and Anthropic both easily exceed the 2025 revenue thresholds, and in fact both have presumably already exceeded the 2026 revenue thresholds! Even if a solar storm knocked out all electronics for the rest of the year I think Marcus would win because OpenAI and Anthropic have already exceeded the revenue criteria for both 2025 and 2026, and Nvidia alone missing its quarterly criteria isn’t enough for Remmelt to win.
OpenAI: according to CNBC they had $13.1 billion in revenue in 2025, according to their blog their ARR at the end of 2025 was $20B, and according to The Information their ARR as of February 2026 was $25B. Even if we use the lower end and assume their average revenue in January and February was $20B annualized, they’ve already made $3.3B in revenue in 2026, meaning they’ve surpassed both revenue thresholds.
Anthropic: according to their blog they had an ARR of $1B in January 2025, and by February 2026 it’s up to $14B. According to the LA Times their ARR was $9B at the end of 2025. Again, even if we use the lower numbers (estimating $1B in revenue in 2025 and $1.5B so far in 2026) they’ve easily hit their targets.