The Data Exchange with Ben Lorica From Preparation to Recovery: Mastering AI Incident Response
Jul 18, 2024
Andrew Burt, co-founder of Luminos.Law and Luminos.ai, discusses AI incident response challenges and preparation. Topics include defining incidents in AI systems, specialized response teams, regulations like SB 1047, contrasting US and European approaches to AI regulation, and the importance of detecting and stopping AI failures.
AI Snips
Chapters
Transcript
Episode notes
Google's Gorilla Mislabeling Incident
- In 2016, Google's photo object recognition mislabeled Black people as gorillas.
- This incident, discovered on social media, highlights the need for AI incident response preparation.
Define AI Incidents
- Define what constitutes an AI incident for your company.
- Consider unintended behavior causing harm to users or the company as a starting point.
AI's Inherent Imperfection
- AI systems, being probabilistic, will always have some degree of error.
- This inherent imperfection necessitates preparing for incidents, unlike traditional software.