Don't forget about how the presentations for Google Gemini and OpenAI Sora and Devin were all ~~faked~~ embellished too, but no one talks about that anymore either.
I'm still waiting for even one actual use-case for sexy Clippy that isn't generating SEO spam while speedrunning climate change.
And we can do all of that by just scaling up autocomplete which is basically already AGI (if you squint).
But it shouldn't surprise me that their benchmark of intelligence is basically that something can put together somewhat coherent sounding technobabble while being unable to do something my five year-old kindergartner can.
Yup, basically AGI.