There were eight members at the meeting. There was a wide-ranging discussion.

Peter Whitham introduced several papers briefly.

Certificates in AI: Learn but Verify.

https://cacm.acm.org/research/certificates-in-ai-learn-but-verify/ is an important article about ensuring AI solutions meet a requirements specification. An introductory video on this page is a useful summary.

From Model Training to Model Raising

https://cacm.acm.org/opinion/from-model-training-to-model-raising proposes a way to reform AI model-training paradigms from post hoc alignment to intrinsic, identity-based development.

Not presented at the meeting, but well worth reading....

Against Imaginary Friends: Why Digital Companions Are No Solution to Social Isolation

https://cacm.acm.org/research/against-imaginary-friends-why-digital-companions-are-no-solution-to-social-isolation Again, the introductory video on this page is strongly recommended.