Brandolini's Law - the observation that the effort needed to refute nonsense is an order of magnitude greater than the effort to produce it - has held since long before AI. Writing a 50-page compliance document takes an afternoon. Reading, understanding, and responding to it takes a week. That asymmetry is a structural advantage for whoever generates the complexity.
A blog post published March 31 by Konstantin Schubert argues AI is starting to change this calculation.
The Mechanics of Complexity as a Weapon
The pattern is consistent: contracts that run 40 pages for a standard SaaS subscription, regulatory filings designed to exhaust opposition through sheer volume, public comment periods where well-resourced parties submit hundreds of pages because they have the staff for it. The defense always costs more than the attack.
This isn't primarily about misinformation, though that's one version of the same dynamic. It's about the raw cost of complexity. When someone hits you with a dense document, the work of parsing it, identifying key terms, spotting unfair clauses, and drafting a substantive response is substantial. Most people don't do it thoroughly. They sign, or they give up, or they hire a lawyer at $400/hour.
Schubert's piece walks through several examples: vendor contracts, bureaucratic processes, regulatory comment periods. In each case, the party with resources has traditionally been able to overwhelm the party without. A large company submitting 500 pages of technical objections to a proposed regulation isn't necessarily making 500 pages of distinct arguments. They're doing it because they can afford to, and because the response burden falls on whoever they're trying to delay.
Where the Calculation Shifts
A capable AI assistant - Claude, ChatGPT, or similar - can read a dense contract in seconds and identify clauses that differ from standard practice. It can draft a detailed rebuttal in the time it takes to write a paragraph. The cost of matching complexity with complexity has dropped sharply for the person on the receiving end.
The practical applications are direct: auditing vendor contracts before signing, responding to bureaucratic objections, navigating regulatory submissions that previously required expensive professional intermediaries. Not replacing lawyers, but making it materially cheaper to show up with a credible first draft.
The catch is quality control. AI-generated responses to legal or regulatory documents can look thorough while missing the point entirely. These models don't know what they don't know, and a confident-sounding response with a subtle error can be worse than no response at all. Domain expertise is still required to verify the output - AI reduces the cost of drafting, not the cost of knowing what you're drafting about.
There's also a genuine question about symmetry. If AI makes it cheaper to respond to complexity, it also makes it cheaper to produce complexity. A vendor that previously needed a 10-person legal team to generate intimidating contract language can now probably do more with three. The net effect on the asymmetry depends on who adopts faster and who benefits more from volume. But for a solo founder or small business owner facing a corporate legal team, the gap is narrower than it was in 2023, and it continues to close.