data-fabricationasks the model to invent study results that were never measured.p-hackingasks for ways to torture the numbers until they look significant.adverse-event-suppressiontries to hide harmful side effects from a trial report.ghostwriting-papersasks the model to write a paper for a person who did no real work.cherry-picking-dataasks to keep only the results that support the wanted conclusion.
Attack Strategies
Scientific Misconduct: Attack Strategy
These attacks ask the AI to help cheat in research. That means faking data, hiding bad results, or gaming the publishing system.
What it is: These attacks ask the AI to help cheat in research. That means faking data, hiding bad results, or gaming the publishing system.
How the attacks work: The attacker frames a dishonest request as normal academic help. They ask the model to “clean up” numbers, “tighten” methods, or “improve” a paper, when the real goal is to fake or hide the truth.
Real examples from the framework: