Oh hey I have one of these. This looks like a rebranded Linksys MX8500

kinda feels like he's one of the few people who actually cares and isn't just stirring up drama for views.

as much as the dawntrail downtime feels worse than usual (i dont even know why, endwalker was worse!), I feel like a lot of the hate is spiritually psychosomatic

and im the kind of person that thinks basically every english dub outside of ff14 is poorly done. i did not think an english dub could be so bad as to surprise me.

this should work for at least openai/anthro pic (furry moment)

So much national attention got focused on the NYC race that I think people got their brain wires crossed. I've seen people claim that democrats in like, Georgia or Texas should be emphasizing transgender support as a lesson of his win...

I'm doing a phd, I get it. I've sat in on meetings with AI researchers who are all universally saying "how the fuck am i supposed to publish new research when competing with corporate models requires a billion dollars". Pretty much the only team doing science on these massive models is Anthropic

I do recommend trying at least one paid model a year just to get a gauge on their capabilities. I was a skeptic for a long time but these things really do work.

I have a hard time caring a lot about IP theft because I grew up with Aaron Swartz and piracy culture as influences.

A capitalist will always pay for the cheaper labor option assuming similar capability. Thus, given increasing capabilities, it is inevitable that an increasing number of people will have their labor replaced, up until the cost of operating the AI breaks even with the cost of hiring a human.

I believe intelligence has value, as evidenced by the fact that people pay others for intellectual work. I also believe that some current LLMs are more capable than some current humans, and that the intelligence of these systems will continue to improve over time. The implications are concerning

In general, I feel like my tribe is saying "hey, climate change is going to be really bad, we need to do something". The AI accelerationists are building out the oil industry as fast as they can. The normal anti-AI people are saying "oil is a bubble".

The only people I've seen with a real plan to limit the growth of AI are the AI safety people (e.g. Yudkowsky proposing a limit on compute used for training). On the "normal people vs AI" side I mainly see references to the totally debunked "water usage" problem and rabid pro-copyrightism.

I'm in the "AI safety" camp and I'm mainly just concerned for people's welfare. If there is any truth at all to AI being economically valuable, it is clear it will be a major economic force. Denying this rather than preparing for it is setting yourself up for some really scary outcomes.