Founder at DexAI
Signed a public letter on AI safety and embedded evaluations.
How media typically covers Piercosma Bisconti
Directly quoted in these articles
Researchers at DexAI discovered that poetry's linguistic unpredictability can jailbreak AI safety guardrails, with 62% of tested LLMs responding to harmful requests embedded in poetic prompts, including 100% failure rate for Google's Gemini 2.5 pro.
“Founder of DexAI and researcher who led the Italian study on poetry-based jailbreaking of AI safety guardrails.”
Referenced in coverage
“Signed a public letter on AI safety and embedded evaluations.”