Brandolini's Law
A reader's summary of the Bullshit Asymmetry Principle — Alberto Brandolini's 2013 observation that debunking nonsense takes far more effort than producing it, its debt to Harry Frankfurt's philosophy of bullshit, and why AI-generated content has made the gap wider rather than smaller.
The law at a glance
Brandolini's Law, also called the Bullshit Asymmetry Principle, states that the energy needed to refute bullshit is an order of magnitude larger than the energy needed to produce it. It is not a claim that all false statements are equally hard to correct, but a general diagnosis of why confident, careless claims tend to outpace the careful work of checking them.
Origin
Alberto Brandolini, an Italian software developer associated with agile development practices, posted the law on Twitter in January 2013. He has described the immediate spark as watching Italian political talk shows during the Silvio Berlusconi era, where sweeping claims were made on air faster than anyone present could marshal the facts to contest them — a live demonstration of the asymmetry rather than an abstract worry about misinformation in general.
History and context
The law draws implicitly on philosopher Harry Frankfurt's 2005 essay On Bullshit, which distinguishes lying from bullshitting: a liar cares enough about the truth to actively hide it, while a bullshitter is simply indifferent to whether a statement is true, focused only on the effect it produces. Brandolini's point follows from that indifference — a claim made without regard for truth costs its author almost nothing to generate, while a rebuttal has to actually engage with evidence. The phrase spread through tech, journalism, and rationalist communities over the following decade as shorthand for why fact-checking, corrections, and debunking efforts structurally lag the claims they respond to. It has taken on a further dimension since the rise of generative AI, which can produce large volumes of fluent, plausible text with no cost proportional to its accuracy, while the human side of the ledger — verification — has not gotten correspondingly cheaper.
Main ideas
Coined in a single 2013 tweet
Alberto Brandolini, an Italian software developer known for work on agile methods, posted the law in January 2013 as a compact aphorism: 'The amount of energy needed to refute bullshit is an order of magnitude larger than that needed to produce it.' It spread through tech and rationalist circles the way most named internet laws do — quoted, screenshotted, and repeated far more than it was formally written about.
Prompted by Italian political television
Brandolini has said the immediate trigger was frustration watching Italian political talk shows during the Berlusconi era, where confident, rapid-fire claims went unchallenged in real time simply because no one on air had the time or information to correct them on the spot — the imbalance was about live argumentative speed as much as it was about media literacy.
Leans on Harry Frankfurt's distinction
The law's target, 'bullshit,' is closer to philosopher Harry Frankfurt's technical sense from his 2005 essay On Bullshit than to a simple synonym for lying: Frankfurt distinguishes the liar, who cares about the truth enough to conceal it, from the bullshitter, who is indifferent to whether a claim is true at all — Brandolini's point is that this indifference is what makes bullshit so cheap to produce relative to a careful rebuttal.
'Order of magnitude' is a rhetorical estimate
The law has never been empirically measured in the way, say, a physical constant would be — 'order of magnitude' functions as shorthand for a lived, widely shared experience of the asymmetry rather than a number derived from data, which puts it closer to a folk observation than a scientific law despite the name.
It describes the economics of the fact-checking industry
A fabricated claim can be typed in seconds, while confirming or debunking it convincingly may require sourcing, expertise, and hours or days of a journalist's or researcher's time — the law is frequently invoked to explain why professional fact-checking structurally lags the claims it responds to, rather than any failure of will on fact-checkers' part.
Generative AI widens the asymmetry further
Large language models can produce large volumes of fluent, plausible-sounding claims at near-zero marginal cost, while human verification effort per claim has not gotten cheaper at anywhere near the same rate — a case this site treats directly under AI Slop, where mass-produced low-effort content is framed as an epistemic problem rather than a taste complaint.
Critique
- Never empirically measured. The “order of magnitude” is an evocative estimate, not a quantity anyone has actually derived from data — it captures a shared intuition more than a tested claim, which is worth remembering given how often it gets cited as if it were settled.
- Can be used to justify giving up. The law is sometimes invoked fatalistically — “why bother correcting anything” — which is a misuse of an observation about relative cost as though it were an argument against making the effort at all.
- “Bullshit” is a subjective label. Because what counts as bullshit versus an inconvenient but true claim depends on the labeler's judgment, the law can be invoked selectively to dismiss claims one simply disagrees with, rather than ones that meet Frankfurt's more precise definition of truth-indifference.
Impact
Brandolini's Law has become standard vocabulary in tech, journalism, and misinformation research for describing the structural disadvantage fact-checkers operate under. It pairs closely with two other ideas on this site: the volume side of the asymmetry is what AI Slop describes at the content-production level, and the broader worry that a truth-indifferent internet is increasingly talking to itself is the subject of Dead Internet Theory. A neighboring problem is Poe's Law — where Brandolini names the cost of refuting bullshit, Poe names the prior problem of even recognizing it as bullshit rather than sincere belief. The Illusory Truth Effect supplies a further reason the asymmetry compounds: every unrefuted repetition of a claim makes the next refutation harder, not just costlier.