| domain | kindo.dev |
| summary | This public talk explores an unconventional application of classical rhetoric in the context of generative AI models, aiming to bypass safety guardrails that typically protect against misuse or harmful outputs. The presentation intertwines ancient philosophical techniques with contemporary AI security concerns, highlighting the potential manipulation of AI systems through persuasive language and argumentation.
The speaker possesses 9 years of professional engagement in AI ethics, spanning from contributions to IEEE standards development to conducting red-teaming exercises designed to identify vulnerabilities in AI systems. This background positions the speaker as uniquely qualified to address both the theoretical underpinnings and practical implications of integrating rhetoric with AI security measures.
The talk likely serves as a thought provocation for attendees, encouraging critical reflection on how time-honored methods of persuasion might be exploited in the digital age to circumvent safety mechanisms in AI technology. The aim is to stimulate dialogue around the intersection of history and current technological challenges, prompting a reevaluation of AI ethics governance frameworks. |
| title | Gerol Petruzella, Ph.D. | Philosopher-Technologist | AI Ethics |
| description | Gerol Petruzella, Ph.D. - Philosopher-technologist bridging humanities research and applied AI ethics. Academic Technology Consultant at Williams College, AI ethics researcher, speaker, and consultant. |
| keywords | ethics, technology, work, years, academic, view, technical, williams, college, philosophy, risks, consultant, author, education, ancient, nist, contact |
| upstreams |
|
| downstreams |
|
| nslookup | A 199.36.158.100 |
| created | 2025-11-08 |
| updated | 2025-11-08 |
| summarized | None |
|
|