Antifragility
Antifragility is the property of things that gain from disorder — they grow stronger under stress, volatility, and shocks rather than merely resisting them. It goes beyond resilience: the resilient survive chaos; the antifragile improve because of it.
How it works
Some systems have asymmetric exposure to randomness — small stressors trigger adaptation and overcompensation. Muscles, immune systems, and decentralised markets strengthen from manageable shocks, provided no single shock is fatal.
How to use it
- Build in small, frequent stressors plus recovery rather than avoiding all stress — like training a muscle.
- Cap the downside so no shock is fatal, then let volatility do its work (the barbell strategy).
- Prefer options that benefit from surprise over ones that merely survive it.
Worked example
Muscles weaken without stress and strengthen under progressive load plus rest. A startup that learns fast from small failures becomes antifragile; one built merely to be shock-proof stays static.
Where it fails
Antifragility needs bounded, survivable stressors — a fatal shock kills before any gain. It’s not “embrace all chaos”; it’s “expose to small shocks, cap the catastrophic ones.”
The deeper point
Robust and antifragile look identical in calm weather — the difference only shows under stress. Most things marketed as "stable" are actually fragile: they’ve simply hidden their volatility until it arrives all at once.
Frequently asked
- What is antifragility?
- The quality of getting stronger from stress and disorder — not just surviving shocks (resilience) but improving because of them, like muscles under load.
- What’s the difference between resilient and antifragile?
- Resilient things resist shocks and stay the same; antifragile things gain from shocks and get better — as long as no single shock is fatal.
- Who coined antifragility?
- Nassim Nicholas Taleb, in his 2012 book Antifragile, extending his earlier work on randomness and risk.
Related
Editorial synthesis © ReadGlobe 2026, drawing on the mental-models tradition (Charlie Munger, Farnam Street) and the primary sources for each model. · Last reviewed 2026-05-29.