This is a linkpost for https://blog.google/technology/ai/introducing-pathways-next-generation-ai-architecture/
Sounds like they're planning to build a multimodal transformer. Which isn't surprising, given that Facebook and OpenAI are working on in this as well. Think of this as Google's version of GPT-4.
I'm firmly in the "GPT-N is not AGI" camp, but opinions vary regarding this particular point.
Seems like a model capable of generalisation.
And a multimodal one, too.
Anyone got more info or demo of that? They seem to claim a lot, but don't have anything to show yet; it's not clear to me why they would release an announcement so abstract. More to come, I guess?
Seems like they have an architecture but they are yet to build on it. And they don't share any details of the architecture; maybe they consider it a memetic hazard?
This all sounds very concerning and checks many of the boxes of potential true AGI for me.
What are your thoughts?