What behavioral indicators reliably signal attempts to game deliberative processes?

Google’s guide defining adversarial testing as systematically evaluating ML models against malicious or inadvertently harmful input, covering explicit queries (containing policy-violating language) and implicit queries (seeming harmless but involving sensitive topics). The four-stage workflow inv...

Satterthwaite’s landmark 1975 work on strategy-proofness and Arrow’s conditions, investigating the relationship between preventing strategic manipulation in voting procedures and satisfying Arrow’s impossibility conditions. This foundational work in mechanism design theory demonstrates existence ...

Hardt et al. (2015) address classifier manipulation by strategic actors, modeling the problem as a sequential game between classifier designers and individuals seeking favorable classification who may alter attributes to game the system. For natural cost function classes, they developed computati...

Miller, Milli, and Hardt (2020) reveal a fundamental connection between strategic classification and causal inference, distinguishing between gaming (circumventing the system) and genuine improvement. Their central argument is that designing classifiers that incentivize improvement must inevitabl...