disgruntledphd2

7 ms·

91.
by disgruntledphd2 2mo ago
To quote a 17c vulgarity, reason is a whore.
92.
by disgruntledphd2 2mo ago
GPT is less bad for this, which is why I've mostly shifted to using it.
93.
by disgruntledphd2 2mo ago
> It's every company. If you reject companies for this, you are living under a bridge right now. I just finished a successful job search, and only had 1 company (of maybe 50 applications) try this.
94.
by disgruntledphd2 2mo ago
> Well then it's not financially worth it to have a job, because every company does this because if they didn't they'd need to triple in size with recruiters. Every company does not do this. I always refuse tasks like th
95.
by disgruntledphd2 2mo ago
> But, in US, people are so alarmed about US' economy collapsing. It's more than US Treasuries (debt) are the safe asset upon which all other assets are priced. If they go mental, then lots of assumptions break and the machines
96.
by disgruntledphd2 2mo ago
> Inference is profitable Let's wait for Anthropic's S-1 to see if this is actually true. Fundamentally, if Anthropic ain't profitable with all the investment and usage they get, then it's possible that it may not be
97.
by disgruntledphd2 2mo ago
Same model, but they're more like a holding company for a bunch of tech stuff. They're apparently a great employer in Italy, which is a much better cost structure than the US (from their perspective).
98.
by disgruntledphd2 2mo ago
> The fact the Anthropic announced that their own model did basically the same thing within a couple of weeks does, to me, suggest their is a strong PR driver to all this. A more likely explanation is that RL training incentivises basica
99.
by disgruntledphd2 2mo ago
Anthropic are definitely profitable on my 20 eur subscription, as I have kids and no time.
100.
by disgruntledphd2 2mo ago
> Big enterprises are on a different github deployment, so I doubt they are affected by this Depends, some of them use GH Enterprise but many (surprisingly) don't.
101.
by disgruntledphd2 2mo ago
This plan would still take 3-5 years, even in the best case scenario. Clinical trials are really hard.
102.
by disgruntledphd2 2mo ago
And hands.
103.
by disgruntledphd2 2mo ago
> Drug discovery is similar AFAIK. The space of possibilities is even larger than protein folding, but it's structurally similar enough that I think AI will help to make progress on the discovery side. I do agree that this kind of t
104.
by disgruntledphd2 2mo ago
Yeah, it's particularly common when the EU does stuff.
105.
by disgruntledphd2 2mo ago
> I mean they did solve the protein folding problem did they? They made huge progress, but I would say that the vast majority of work on this problem was designing the harness for the model. That's a lot of work for each and every d
106.
by disgruntledphd2 2mo ago
> Curing cancer sounds insane, but it's also a research problem, I just don't buy the AI labs approach to this stuff. Like, unless we can basically simulate the entirety of human biology, I don't really see how LLMs can ma
107.
by disgruntledphd2 2mo ago
This also happened in 2018, but yeah it's crazy.
108.
by disgruntledphd2 2mo ago
Ireland took in an incredibly large number of Ukraine refugees per capita. The alumina thing is pretty bad, and reflective of Irish peoples relatively confused understanding of geopolitics. Now do the Greeks ;) one of the big problems with
109.
by disgruntledphd2 2mo ago
Crash means zero reward, so yeah.
110.
by disgruntledphd2 2mo ago
I mean to be fair to him, he's built an innovative car company, a pretty rocking space company, and now controls access to basically all government officials and politicians (X). He's definitely super successful and not a complete
111.
by disgruntledphd2 2mo ago
For me, at least it's that the newer Claude models seem optimised for one-shotting things, which is not what I want. As the amount of code per turn increases, I have a harder job keeping up and ensuring that it's doing what I want
112.
by disgruntledphd2 2mo ago
Yeah, I actually have started using GPT Sol much much more, as Claude (all of them) were far too trigger happy around making changes, and refused to listen to my requests to take things slowly. Feels like they've overtrained on one-sho
113.
by disgruntledphd2 2mo ago
> I would avoid phrasing it as "better". Less wealthy parents are likely working more hours, have to spend more time on household maintenance (don't have a cleaner, can't afford grocery delivery), not to mention the o
114.
by disgruntledphd2 2mo ago
Mine told me that this always happens, so I schedule my appointments appropriately.
115.
by disgruntledphd2 2mo ago
> The benefit richer students have is that, on average, their parents are smarter, more conscientious, etc. Yup. In general the benefit of schools in richer areas is better parents, rather than better schooling itself. Depressing, but tr
116.
by disgruntledphd2 2mo ago
> The doctor would give appointment times that assumed he'd spend 15 minutes per patients but he was friendly and patient with answering questions so he'd always end up being very late by the time evening came. I have a doctor
117.
by disgruntledphd2 2mo ago
> That's magnesium deficiency. Promise. I find that incredibly hard to believe (although my doctor did mention that magnesium supplements were worth taking). Can you point me to some research that support this claim please?
118.
by disgruntledphd2 2mo ago
I don't really, but at least there would be standards. Like, the massive amount of hacks and breaches we've seen in the past ten years indicate that something needs to happen.
119.
by disgruntledphd2 2mo ago
> Once again the alternative is not nothing. Can you help me understand what the alternative is?
120.
by disgruntledphd2 2mo ago
> As I commented elsewhere - the problem then becomes who gets to define what the proper path is. For example destroying companies because they chose not to follow the latest sage advice from anyone who once signed the Agile Manifesto do

More ›