{"slug": "when-the-downside-is-limited-and-the-upside-is-not-variance-is-your-friend-when", "title": "When the downside is limited and the upside is not, variance is your friend. When the downside is possibly very negative, it is not.", "summary": "When the downside of a donation is small and bounded but the upside is very high, variance becomes beneficial for donors, as only a few winners are needed for a great outcome. However, when the downside is potentially very negative, such as with AI safety research that could lead to more dangerous AI or political lobbying that could backfire, variance is harmful. This asymmetry suggests donors should favor uncertain global health interventions over longtermist causes with less direct theories of change.", "body_md": "*I used an LLM to help review this post and it likely contains some AI-generated re-formulations. The ideas are not fundametally new and inspired by Nassim Taleb and trading lore.*\n\nIn effective giving, it makes sense to be close to risk-neutral and focus on high expected utility per dollar.\n\nHowever, in practice, organisations and people are risk-averse for good reasons.\n\nIt can add value to take more risk as an individual donor to generate higher expected utility at the margin. But when?\n\nOne important angle that I increasingly think matters a lot more than commonly appreciated is the following: If the downside is small and bounded, but the potential upside is very high, variance is good (cf. venture investing, convex macro trades). If you add diversification, you just need a few winners to have a good chance of realising a great outcome.\n\nIf the distribution of possible outcomes is close to symmetrical, variance either does not matter (only the mean matters). If there is a non-linearity of utility on the downside, variance is bad (selling convexity).\n\nThis line of reasoning makes it much easier to donate to uncertain global health interventions that might not work, but are unlikely to cause harm. However, it makes it harder to donate to longtermist cause areas where the theory of change is less direct, such as AI safety research which might lead to more dangerous AI, or to political lobbying that could really backfire for the organisations involved.", "url": "https://wpnews.pro/news/when-the-downside-is-limited-and-the-upside-is-not-variance-is-your-friend-when", "canonical_source": "https://forum.effectivealtruism.org/posts/vRpGjXouoDCEAeLmh/when-the-downside-is-limited-and-the-upside-is-not-variance", "published_at": "2026-06-06 18:37:44+00:00", "updated_at": "2026-06-06 19:32:31.120747+00:00", "lang": "en", "topics": ["ai-safety", "ai-research", "ai-ethics"], "entities": ["Nassim Taleb"], "alternates": {"html": "https://wpnews.pro/news/when-the-downside-is-limited-and-the-upside-is-not-variance-is-your-friend-when", "markdown": "https://wpnews.pro/news/when-the-downside-is-limited-and-the-upside-is-not-variance-is-your-friend-when.md", "text": "https://wpnews.pro/news/when-the-downside-is-limited-and-the-upside-is-not-variance-is-your-friend-when.txt", "jsonld": "https://wpnews.pro/news/when-the-downside-is-limited-and-the-upside-is-not-variance-is-your-friend-when.jsonld"}}