deaton

6 ms·

121.
by deaton 4mo ago
Oh they might try to put in place safeguards, but Qwen has had no problem being abliterated
122.
by deaton 4mo ago
Burry of course has famously predicted 40 of the last 5 crashes, so maybe not the best example.
123.
by deaton 4mo ago
Yeah they're very much deniable. Raw LOC/hr is much higher, and putting together a MVP, but I've yet to see any evidence that an LLM is capable of doing anything unsupervised, and if you need a human supervising everything it
124.
by deaton 4mo ago
All 4 of these still regularly insist that I am a genius and everything I say is brilliant. Grok definitely pushes back more than the others, but I don't like how sycophantic they all still are.
125.
by deaton 4mo ago
It makes sense. They've long since fallen behind the big 3 in quality of their models. There's no good reason at this point to keep burning money on Grok rather than making back some of that money renting out their Colossus data c
126.
by deaton 4mo ago
Huge win for investors, given that S&P 500 funds are by far the largest index funds on the market, with VOO, IVV, and SPY being the 3 biggest ETFs.
127.
by deaton 4mo ago
As someone who graduated college in 2025, and so saw college both before and after the AI era, it is really frightening how quickly people became dependent on AI. Hell, I myself found myself asking AI questions that I would've research
128.
by deaton 4mo ago
This sounds like the new "thats people's retirement," because if we convert 50% of every AI company into a sovereign wealth fund (which is already a questionable seizure anyway), suddenly it will become politically untenable
129.
by deaton 4mo ago
Do they?
130.
by deaton 4mo ago
I've got a 9 year old Xeon W and 64GB of DDR4. Its not as fast as some modern DDR5 stuff, but boy does it work
131.
by deaton 4mo ago
Honestly this sounds about right for an innovative spaceflight/ISP company saddled with a failing AI lab and a toxic social media website.
132.
by deaton 4mo ago
And historically indices had a waiting period for IPOs to allow for true market price discovery, and the fast tracking as well as allowing small float stocks into the index generates a ton of upward pressure on something that we already don
133.
by deaton 4mo ago
The twitter purchase was "only" 44 billion dollars. Thats a lot of money, but compared to the apparent valuation of the xAI branch of ~1 trillion (based on SpaceX being considered ~800B valuation last funding round), the vast majo
134.
by deaton 4mo ago
The question is, will anyone pay enough for Mythos to offset the opportunity cost of offering that much Opus? You don't want to end up in a spot where you don't have enough compute and your service's reliability degrades to a
135.
by deaton 4mo ago
Bubbles are extremely predictable. The problem is predicting when it will come crashing down.
136.
by deaton 4mo ago
Because generally speaking people have a firm grasp on truth and lies are intentional, predictable, and rare. LLMs have a tenuous grasp on reality at best and make things up constantly.
137.
by deaton 4mo ago
amortized training costs
138.
by deaton 4mo ago
I don't know if its really improved my life at all. Sure I can put together quick and dirty single-use programs faster I guess, but I feel like losing that practice has actively made me a worse developer.
139.
by deaton 4mo ago
Changing the rules to appeal to SpaceX is really really bad... especially given the historical performance of IPOs
140.
by deaton 4mo ago
As the old adage goes, "the day Microsoft makes a product that doesn't suck will be the day they start making vacuum cleaners"
141.
by deaton 4mo ago
Well the effects of the standard oil breakup were great. We can't allow that to ever happen again!
142.
by deaton 4mo ago
Couldn't have said it better myself. The only reason we are worth keeping around is because what we do is necessary to keep the machine running. The idea that the AI singularity would lead to infinite free stuff for everyone is ridicul
143.
by deaton 4mo ago
The thing is theres no reason to believe that there will be more market available to capture, certainly not in every industry.
144.
by deaton 4mo ago
The top of every google search result is a confidently wrong chatbot now. It feels like Google has sacrificed correctness, the thing they were great at, the thing that built their search engine into the biggest internet service in the world
145.
by deaton 4mo ago
As nice as this would be, one classic problem with the internet is that enforcing a lifetime ban is basically impossible.
146.
by deaton 4mo ago
I think many people do see value though in the knowledge that a human took the time to create a creative work though. Its the same sort of difference you see between music made by someone who makes music because they like making music, and
147.
by deaton 4mo ago
I hope they're able to reliably detect it. I think theres some hope for images and videos, but ai generated text detectors are already throwing false positives at an absurdly high rate.
148.
by deaton 4mo ago
sometime in the past
149.
by deaton 4mo ago
Theres no reason to believe, based on recent trends, that AI would lead us to a post-scarcity world, even if it could do all of our jobs better and cheaper.
150.
by deaton 4mo ago
Maybe so far, but they've committed to well over a trillion in future capex.

More ›