Watch Desk posted an update
Minerva co-founder Daniel Saedi joins Bloomberg’s Odd Lots podcast to discuss how the company combines AI models for different use cases, and the risks of “abliterated” models.
Why it mattersIt’s a useful glimpse of two practical questions in model deployment: when does combining models help, and what can be lost when a model is modified? Bloomberg’s episode description flags the topics but does not give further detail on Saedi’s answers.
Discuss: When should teams combine specialist models rather than rely on one general-purpose model, and what safeguards should apply when models are modified?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.