[artninja1988]

My starting point for a human friendly AGI benchmark is Commander Data from the Starship Enterprise.

On a more thoughtful note perhaps, I believe AGI would have to emulate the human capability of changing the attentional conductor focus (see cognitive architecture chart) based on sensors that bring the attentional element and its aperture onto emergent elements or signals. This would then have to pull the appropriate elements into the simulation space for computation. In the space of LLMs I believe the transformer functions are pretty limited in scope for attentional context. I would be interested to hear how others think about the (technical) evolution to AGI at high level and whether that would include the social signals and behavioral mediation that characterize human interactions and intelligence.

[creer]

Re. transformer "attention". Might this not be a higher-level function? i mean it could be both but for example, monitoring and being alert to outside stimulus might be a higher level trait. One that becomes part of planning and goal-forming erm, architecture. Not necessarily one that's part of the language base "assimilation".

There might be language base separately from incoming streams.

[rakkhi]

Who is right? Google: LLMs are already AGI

Or me: LLMs are just useful for experts:

https://open.substack.com/pub/rakkhi/p/economics-of-large-la...

[creer]

The fine article addresses most of this. In their terms, you are complaining that this ENIAC of AGI is not performing like a CAD workstation or something. Of course it isn't. The question is not how performant it is, but whether it is AGI - no matter how insufficient for our ideal Jetsonian needs.

Your only argument for it not being AGI is "clearly". You'll need to expand on that part.