# When AI Ignores the Word "Not": Difficulty With Negation, Mode 22 | IgnatiusTheYoungerAI > Difficulty with negation is a model's unreliable handling of negating words, where one token reverses meaning while barely shifting the statistical signal. - canonical: https://ignatiustheyoungerai.com/failure-modes/22 - html: https://ignatiustheyoungerai.com/failure-modes/22 - machine-view: https://ignatiustheyoungerai.com/failure-modes/22?view=machine - markdown: https://ignatiustheyoungerai.com/failure-modes/22.md - site-index: https://ignatiustheyoungerai.com/llms.txt - evidence-labels: RESEARCH · REPORTED · ANALYSIS · MODELED · RECOMMENDATION · OBSERVED — carry the label with the claim; MODELED is arithmetic from stated assumptions, never a measured outcome - publisher: IgnatiusTheYoungerAI (Nathan Battin) · hello@ignatiustheyoungerai.com --- Failure Mode 22 of 24 # Difficulty With Negation > 'Must not' and 'must' differ by one word. That word is frequently the whole control. Negation is a small lexical signal carrying a total semantic reversal. In distributional terms, "the report may be shared" and "the report may not be shared" are extremely similar strings, and the training signal for handling the difference correctly is weaker than the frequency of the surrounding pattern. By [IgnatiusTheYoungerAI](/about) · Last reviewed 2026-07-30 · Judgment Multiple ~19x to ~385x (modeled) · From Part II of the AI "Keep Your Career" Bible In plain English **This page covers one specific way AI gets things wrong at work** , and what to do about it. It runs in order. What goes wrong, why it happens, where you'd notice it on an ordinary day, who takes the blame, roughly what it costs, and the check that catches it. Then one thing to try this week. The dollar figures are estimates, not measurements. The assumptions behind each one are printed right there, so you can swap in numbers that fit your job. Anything actually measured carries an OBSERVED tag. ## What is difficulty with negation? Negation is a small lexical signal carrying a total semantic reversal. In distributional terms, "the report may be shared" and "the report may not be shared" are extremely similar strings, and the training signal for handling the difference correctly is weaker than the frequency of the surrounding pattern. This compounds in three ways: 1. **Scope ambiguity.** What exactly does the negation cover? "Employees may not use personal devices to access customer data outside secure environments." The negation's scope is genuinely unclear even to careful human readers. 2. **Double negation and exceptions.** "Except where prohibited" constructions invert meaning again. 3. **Summarization loss.** When condensing, negations are the most frequently dropped element, because they're syntactically small and carry no topical weight. That third one is the operational killer. **A summary of a policy is disproportionately likely to lose the prohibition** , which is the only part of the policy that constrains behavior. ## What do people assume about difficulty with negation? That a system parsing a prohibition understands it as a prohibition, that "may not disclose" and "may disclose" are handled as opposites. Negation is handled less reliably than affirmation, and the failure is silent. ## Where does difficulty with negation show up at work? A team asks an assistant to summarize a data handling policy into a one-page quick reference. The summary is accurate and readable. The clause reading "may not be transferred outside the approved region without written authorization" becomes "may be transferred with authorization." Substantially similar. Operationally inverted. One describes a default prohibition, the other a default permission. ## Who carries the downside? **Vendor:** none. **Executive:** attests to compliance. **Manager:** distributed the quick reference. **You:** produced it. In a regulated transfer, this is a reportable event. ## What does difficulty with negation cost? **[MODELED — not reported]** ``` ASSUMPTIONS Policy / requirement summaries: 40 / year Rate w/ a dropped or inverted prohibition: 10% (~4 / year) Rate caught in review: 50% Bad summaries in circulation: ~2 / year Cost per instance: $10,000 – $200,000 (regulated data incident, remediation, penalty exposure) ``` **Annualized exposure: ~$20,000 – $400,000** ## How do you control for difficulty with negation? Explicit constraint extraction. Before accepting any summary of a policy, contract, or requirement, **list the prohibitions separately** : every "must not," "may not," "shall not," "except," "unless," "prohibited," "restricted." Then verify each one against the source text individually. Prohibitions get their own pass, because they are the part that fails. ``` CONTROL COST Policy summaries: 40 / year Constraint extraction: 20 minutes each Annual: 13 hours Fully loaded rate: $80 / hour ``` **Annualized control cost: $1,040** Judgment Multiple (IgnatiusTheYoungerAI, 2026) — modeled ~19x to ~385x ## What should you do this week? **RECOMMENDATION** Take the policy your team relies on most and extract every prohibition into a standalone list. Compare it against whatever summary or quick-reference your team actually uses day to day. If anything is missing or softened, you have found a live compliance gap **and** produced the artifact that closes it. This is Compliance Mapping (Part III, Concept 21) at individual scale, and it is the highest-stakes twenty minutes in Part II. ## Evidence **RESEARCH** Negation handling as a documented weakness in language models. **ANALYSIS** The claim that summarization disproportionately drops negations is the author's and is the operationally critical assertion in this chapter. The Full System This is one of 24 failure modes. The book gives you all of them, plus the controls that catch each one and a 90-day plan to prove you ran them. [Get the Book · $29](https://agrb01-w9.myshopify.com/cart/43747019161690:1) [← 21 Cross-Session Inconsistency](/failure-modes/21) [23 Cultural And Linguistic Blind Spots →](/failure-modes/23)