Why LLMs Fail (and why AI alignment is needed)

LLMs fail because they're prediction machines, not fact machines. Michelle Robbins, Manager of Strategic Initiatives and Intelligence at LinkedIn, explains why AI models produce inconsistent responses and require critical evaluation. She advocates treating LLMs as thought partners rather than absolute authorities, emphasizing the critical importance of AI alignment to prevent harmful outputs and ensure optimal human outcomes as we progress toward artificial general intelligence.
About the speaker

Michelle Robbins

LinkedIn

 is a little camera shy

Michelle is Manager of Strategic Initiatives and Intelligence at LinkedIn

About the speaker

Michelle Robbins

LinkedIn

 is a little camera shy

Michelle is Manager of Strategic Initiatives and Intelligence at LinkedIn

Up Next: