Today we are thrilled to launch Inside the Incident, a new video series where I talk to Sean McGregor, founder and executive director of the AI Incident Database, about real-life AI incidents in the news.
In our first episode, we tell the story of how one chatbot started selling one dollar Chevy Tahoes - then dissect why it happened, its legal implications, and what AI governance teams can learn from the incident.
Why are we launching this new series?
I’ve been working on AI risks for over a decade - which, like dog years, is hundreds of years in AI time! - and it’s always been hard to separate the substance from the hype. When it comes to AI governance, this trend continues. There’s talk of AI ethics and responsible AI and more frameworks than anyone knows what to do with.
But there’s not enough specificity, and not enough focus on concrete examples that the AI governance community can truly learn from.
That’s why we’ve teamed up with the AI Incident Database, which is doing amazing work to catalogue *actual* examples of AI risks to go beyond the hype and help make conversations about AI governance practical. I can’t emphasize how impressed I’ve been by Sean and his team, and how thrilled I am to be working on this together.
Watch our first episode below - and stay tuned for more!
