alt Hacker NewsAgreed. Another difficulty here is there are not good benchmarks for this new architecture yet, so it’s easy to potshot and snipe, where jev seems to be pretty broadly intelligent/at least have had a lot of rl in different domains.
We haven’t seen any of these copy cats play doom or street fighter for instance; just categorize email.
I imagine once the author cools down and evaluates on a broad harness of tasks he may find that his new thing has a lot of engineering work ahead.
The doom demo would have to be reproduced to confirm what their model is capable of. Oh but it's all closed source, so who knows.
It reminds. Me of Devin. Took a while to debunk. Not saying Jev is a fraud , but the gap between structuring typed output and playing a game involving logical interpretation of frames made of pixels, screams unstructured interpretation they made and forgot to mention.