vonneumannstan

7 ms·

91.
by vonneumannstan 6mo ago
Until recently Claude wasn't building itself. A group of people with agency were.
92.
by vonneumannstan 6mo ago
You can check the hardware costs for self hosting a high end open source model and compare that to the tiers available from the big providers. Pretty hard to believe its not massively subsidized. 2 years of Claude Max costs you 2,400. There
93.
by vonneumannstan 6mo ago
So you are doubly stupid, by not seeing any improvement in the models and also paying for models you believe are terrible? lol
94.
by vonneumannstan 6mo ago
>If we don't innovate, someone else will. Terrible take. You don't get to push the extinction button just because you think China will beat you to the punch. >This is the very nature of being a human being. We summit mountai
95.
by vonneumannstan 6mo ago
Are you guys ready for the bifurcation when the top models are prohibitively expensive to normal users? If your AI budget $2000+ a month? Or are you going to be part of the permanent free tier underclass?
96.
by vonneumannstan 6mo ago
Lol you haven't used a model since GPT2 is what it sounds like.
97.
by vonneumannstan 6mo ago
This seems like something uniquely suited to the startup ecosystem. I.e. offering PQ Encryption Migration as a Service. PQ algorithms exist and now theres a large lift required to get them into the tech with substantial possible value.
98.
by vonneumannstan 7mo ago
Disagree, there are a lot of reasons to use open source local LLMs that aren't related to free/libre/oss principles. Privacy being a major one.
99.
by vonneumannstan 7mo ago
Pretty sure the entire markets for Storage, HBM, DDR5, etc are completely sold out for next several years. How is that saturated?
100.
by vonneumannstan 7mo ago
Do you really need Game Theory to figure out you need to build more houses and can't let NIMBY's be in charge of the decisions for where and when that is done?
101.
by vonneumannstan 7mo ago
>Haven't read his book, but the idea that monopolies are good isn't typically made in a vacuum, it's made relative to alternatives, most often "ham-fisted government intervention". It's easier to take down a
102.
by vonneumannstan 7mo ago
Yeah I mean ChatGPT 5.4 Pro can't even pick my nose for me so it's obviously not AGI /s
103.
by vonneumannstan 7mo ago
Except you will never run out of challenges and my sense from Chollet has been that every challenge was hinted at being the final one where once beaten AGI would have been created and of course at the end of each one he comes out saying aks
104.
by vonneumannstan 7mo ago
>As long as there is a gap between AI and human learning, we do not have AGI. This is an absurd constraint. You could have a vastly superhuman AI that doesn't learn as efficiently as a human and it would not pass this definition whi
105.
by vonneumannstan 7mo ago
>Talking to the ARC folks tonight, it sounds like there will be an ARC-4,5,6,etc. I mean of course there will be. Quintessential goal post moving...
106.
by vonneumannstan 7mo ago
It's getting pretty old now when Francois Chollet puts out a new ARC challenge, claims definitively that no system is going to crack it without being full blown AGI, the benchmark gets saturated in a few months, he claims the systems d
107.
by vonneumannstan 7mo ago
What was the increase in rent across all major cities in the US in 2021?
108.
by vonneumannstan 7mo ago
Ah yes sneaking in via extremely secure international airport checkpoints. Classic
109.
by vonneumannstan 7mo ago
So weird that railing against government overreach is now considered a leftist position when it was a fundamental right position for decades... You could not lick the boot harder if you tried.
110.
by vonneumannstan 7mo ago
This administration has taken grift and corruption to a level only seen in banana republics. I seriously don't know how you come back from this. GOP voters seem to be openly cheering it.
111.
by vonneumannstan 7mo ago
The number of Tech people moving from SF to Austin and back is going to have negligible influence on the average rent in the entire city.
112.
by vonneumannstan 7mo ago
Should be obvious that its tools like Claude Code. If you are a junior dev not experienced in delivering entire products but with good ideas you have incredible leverage now...
113.
by vonneumannstan 7mo ago
Lol most Kuwaitis including the royal family are Sunni and believe Iranian Shia's to be heretics. So no love lost there at all.
114.
by vonneumannstan 7mo ago
Maybe true but you probably expect a minimum of leaked information directly to the CCP if you accept that.
115.
by vonneumannstan 7mo ago
>Can an LLM give you an upfront estimate that a task will take multiple months? >Can it decide intelligently what it would have to change if you said "do what you can to have it ready in half the time?" Do you think ChatGPT
116.
by vonneumannstan 8mo ago
All of which are peanuts compared to Apollo. Not to mention insane cost overruns and timeline failures on their projects like JWST or SLS for example. Many of those successful projects came out of JPL, where I mentioned the talent level is
117.
by vonneumannstan 8mo ago
All fun and games until an Intern deletes the original sheet or fatfingers critical information.
118.
by vonneumannstan 8mo ago
A bit hard for him to swing if he wants to position xAI as a key Defense contractor for AI and his company is full of Chinese Nationals...
119.
by vonneumannstan 8mo ago
Were they kneecapped by Anthropic blocking their distillation attempts?
120.
by vonneumannstan 8mo ago
Task Time horizons are improving exponentially with doubling times around 4 months per METR. At what timescale would you accept that they "can be strategic"? Theres little reason to think they wont be at multi week or month time h

More ›