The Miranda Hypothesis: How Hamilton Poisoned Persona Evals - Jacob E. Thomas, Results Gen
Role-playing persona AI systems are being deployed as civic and pedagogical infrastructure, but persona evaluations may be fundamentally flawed.
“Mine is not better by default. The one thing it is is open. You can read every line of what shapes the persona.”
The speaker argues that role-playing language agents instantiating historical personas (Lincoln, Marcus Aurelius) are moving from entertainment into civic and educational infrastructure, demonstrating how a persona-prompted frontier model produces plausible-sounding but potentially distorted answers on consequential questions. The talk sets up a critique that current benchmarks and evaluations for persona fidelity may be compromised, with implications for how the industry measures and trusts persona instantiation at scale.