All the benchmarks, the comparisons, the GPT vs Claude vs Gemini threads, you opened something, you decided to ask, you had a second to think.
AI is already showing up in other places. Voice agents taking calls, cars making real time decisions, glasses like Ray-Ban Meta, Rokid, XRAI Glass running in the background.
I genuinely don't know how to think about model capability in those environments. I'm not sure the benchmarks were built for it either.
[link] [comments]