banteg's greentext mockery of longtermism sparks discount-rate debate endorsed by philosopher Chappell

TL;DR: On Sept 18, Ethereum developer banteg repeatedly shared a viral greentext mocking how effective altruism slides into longtermism, and added that longtermist moral math collapses once any ordinary discount rate is applied. Philosopher Richard Chappell cited and extended the argument on Sept 19, other users amplified it, and the wave also pulled in criticism of AI safety governance ideas.

Confirmed

• The greentext, originally shared via visakanv and amplified by banteg, narrates in first person a typical EA journey: optimizing donations via spreadsheets (buying bednets), then accepting the moral weight of future people, extending concern to descendants 500, 5,000, even 500 million years out, until the calculator yields 10^50 unborn people in the universe and the current 8 billion become a rounding error, closing with jokes like "10^50 unborn people are the perfect constituency."
• Beyond banteg's multiple reposts, jefrankle and paulnovosad shared the same text on Sept 18; menhguin framed it as a cultural shift "from buying bednets to anxiety about future humanity," showing cross-community spread.
• banteg added two extensions: that longtermism's moral math collapses under an ordinary discount rate, and a link to a "more serious" reading analyzing how EA/longtermist rhetoric gradually recruits people into its doctrine.
• On Sept 19, philosopher Richard Chappell cited banteg's observation and pushed it to a reductio: under ordinary discounting, longtermist conclusions become absurd to the point that we "should torture" future people for a lollipop today (as paraphrased in the post).
• NathanpmYoung's repost of MatthewTanous supplied concrete arithmetic: at a 5% discount rate, people 50 years out are worth 8.7% of people today, 100 years out 0.76%, 200 years out 0.00527%; by 470 years the present value of all humanity falls below today's, arguing for temporal discounting against longtermism.
• Alexandra Barr, author of the quoted original, added that people familiar with EA narratives should remember most people find these views frightening and confusing, and their unease deserves a response.
• banteg also endorsed DrNickA's critique of the AI safety community as trapped in "collective delusions of grandeur," feeling responsible for all humanity and the entire light cone; combined with Benthamite utilitarianism this yields panopticon-style governance—mass surveillance as the price of stopping model distillation or banning open source.

Why it matters

• banteg is an influential crypto developer; his repeated posts turned the greentext into a concentrated Sept 18 wave of mockery that escalated from a meme into substantive philosophical criticism—ordinary temporal discounting breaks longtermist moral aggregation—later endorsed by professional philosopher Chappell and backed by concrete arithmetic. The discussion also surfaced political criticism of AI safety governance proposals (open-source bans, surveillance-style regulation), reflecting long-standing distrust of longtermist narratives in crypto and open-source communities.

2026-09-18 ~ 2026-09-19 · 13 related posts

• [source] Greentext roasts effective altruism: from mosquito nets to galaxies of expected value — banteg · 2026-09-18
• A developer's critique of how EA/longtermism rhetoric pulls people in — banteg · 2026-09-18
• banteg's EA greentext sparks debate: from mosquito nets to 500-million-year futures — banteg · 2026-09-18
• Longtermism critique goes viral: from mosquito nets to 500-million-year math gone absurd — banteg · 2026-09-18
• Viral greentext skewers effective altruism: 10^50 unborn people as the perfect constituency — banteg · 2026-09-18
• Greentext roast captures EA's drift from mosquito nets to longtermism — menhguin · 2026-09-18
• The longtermism greentext: from mosquito nets to calculating humanity's 500-million-year future — paulnovosad · 2026-09-18
• Banteg's greentext nails the EA logic slide: from mosquito nets to donating for the next 500 million years — jefrankle · 2026-09-18
• Debate: Should future people be discounted? Longtermism vs temporal discounting on X — NathanpmYoung · 2026-09-18
• [source] banteg: longtermist moral math 'doesn't survive an ordinary discount rate' — banteg · 2026-09-18
• banteg amplifies critique of AI safety culture: longtermist moral math as 'mass delusion' — banteg · 2026-09-18
• [source] Philosopher RY Chappell: ordinary discount rates break longtermist moral math — AaronBergman18 · 2026-09-19

1 near-duplicate retellings: banteg