But I do get the feeling that personifying agents is something that benefits the bottom line of companies like Anthropic, even while it creates many of the problems they claim to be dealing with.
Anthropic is in the middle of another funding round (when aren't they?) so it is logical that they produce extra bs narrative again.
The kind of stories where coding questions are asked tend to be stories where people provide coding answers.
Yes. Claude models are very well trained on coding adjacent tasks. For that, there is in my opinion no better model; hasn't been for a while. But for generic things, I often find myself on LMArena, letting bots battle out some non-coding, non-critical task, just to gauge what's possible and how good things are getting out of the box, and honestly, Claude is often not doing so well on non-coding tasks.
But then I ask myself: would you ask a coder to run your finance department? Of course not. So it kind of makes sense.
Also: #1415561
Anthropic is in the middle of another funding round (when aren't they?) so it is logical that they produce extra bs narrative again.
Yes. Claude models are very well trained on coding adjacent tasks. For that, there is in my opinion no better model; hasn't been for a while. But for generic things, I often find myself on LMArena, letting bots battle out some non-coding, non-critical task, just to gauge what's possible and how good things are getting out of the box, and honestly, Claude is often not doing so well on non-coding tasks.
But then I ask myself: would you ask a coder to run your finance department? Of course not. So it kind of makes sense.