> OpenAI did the same “too dangerous to release” song and dance for the awesome, world ending AI that was GPT-2.

Wasn't that when Dario, et al were at the company. One way to view this is that OpenAI expelled the cultists and they went on to form their own organization that continued using the same tactics.

Certainly some of the Anthropic press around Fable seems to me to be just marketing but I also think there's a core of people there who really believe it. I also think like all good advertising/lies there's some truth to the claims even if they're exaggerating.

> I find it worrysome how often people value revenge over good.

I personally see it as a net good if companies fearmongering for marketing purposes then have to face consquences from people taking their marketing at face value.

Hopefully it teaches them and others not to do it anymore.

Their entire marketing strategy has been unwarranted fearmongering. This is completely unsurprising.

I personally find the fearmongering annoying and just ignore it, but the model is incredibly capable. It genuinely makes any user of it much more powerful, so to be aware of that is sensible.

As user of course I'd like zero fearmongering, zero regulations, zero drama. This all sucks for us users.

Their coverage area in Orlando is so limited I functionally can't use the service. Would be nice if they could expand it to at least include the airport and tourist areas.

I think all of their metro areas are like this. I'm sure they did some math on the duty cycle of the cars to help draw the boundaries.

They should apologize for their visible gaurdrails, I don't think I've had a conversation that hasn't downgraded to Opus for completely inexplicable reasons.

I'm not an expert in the area but have read a bunch on this topic to try and understand it better. Bird brains and human brains are structured very differently. Birds are much more like GPUs with independent distributed processing happening in parallel. Mammals have these big bidirectional layers where signals are constantly propagating up and down in a big connected computation.

DeepSWE is the benchmark you want to actually look out for. Only one that aligns with actual user reported results from trying the models.

Did you read the blog post? They compare to deepswe and call it out as the worst one for false positives (failed, but the benchmark assessed it as correct). It also has less language variance.

I mean yes that is what you'd say if you were writing a blog post about your new benchmark.

Sure, but they at least quantified it with data. It's not like they just dropped a sentence saying the above, they showed numbers.

There was a point in time when this would have been untrue of Google. They didn't manifest into this market position, they rose there over the years after starting with a relatively modest product offering. Nothing saying someone else can't do the same.

I couldn't get trading working but maybe I'm doing something wrong.