This feels like one of the next big infra layers for AI: tools that can inspect, compare and monitor latent representations across models.
Also strongly relates to @aimalysheva’s point about mapping manifolds and preserving invariants across models.
Reading what AI doesn't say
(this topic genuinely matters to me, and it happens to be my birthday today, so if you feel like it, a repost would be a nice gift 🎂)
Something shifting in AI architecture that I don't think gets discussed enough through a safety lens. Models are
