oofbey

6 ms·

61.
by oofbey 5mo ago
People die either way. People die because treatments sit in the lab for years instead of in clinics. Consider the FLASH oncology treatment. In 1995 Dr Fauvadon figure out how to make radiation therapy much much safer. He spent 14 years befo
62.
by oofbey 5mo ago
Medical research moves. Very. Slowly.
63.
by oofbey 5mo ago
For all the doubt and negativity here I just want to say “good job” to you. Way to take matters into your own hands and protect your love ones. Haters gonna hate but you did it.
64.
by oofbey 5mo ago
GPT is a live example of how LLMs can score very highly on tests and still be a complete moron.
65.
by oofbey 5mo ago
I’m so sorry. American medical institutions are a very long way from the best way to practice medicine.
66.
by oofbey 5mo ago
You’re holding on to the intuition (hope) that we are smarter than the LLMs in some hard to define way. Maybe. But it’s getting harder and harder to define a task that humans beat LLMs on. On pretty much any easily quantifiable test of kno
67.
by oofbey 5mo ago
Nicely done
68.
by oofbey 5mo ago
Wherein OpenAI admits they have very little understanding of how their models’ personality develops. And implicitly admit it’s not all that important to them, except when it gets so out of hand that they get caught making blunt corrections.
69.
by oofbey 5mo ago
True. But that’s never the framing you hear from astronomers. It’s how the satellites are “ruining” the pictures, like this whole thread. The SNR degradation isn’t even very much. Noise goes down by 1/sqrt(N) samples. In a stack like t
70.
by oofbey 5mo ago
70% of all satellites are Starlink. So probably.
71.
by oofbey 5mo ago
Exactly. It’s not that hard to remove the satellites. It’s almost easier than whining about it. But whining is more fun.
72.
by oofbey 5mo ago
Computational photography has long been table stakes for astronomers. They just need to up their game on satellite rejection algorithms. Satellites look nothing like stars, and as such are pretty easy to remove with software. Pictures like
73.
by oofbey 5mo ago
Gemini is perfectly capable of searching the web. Pretty good at it really. As are most agents. If such a surge happens, it’s purely because of laziness.
74.
by oofbey 5mo ago
Mathematical ideas are very difficult to protect. But models can also be improved with brute force improvements in size. Imagine Mythos is a 32 trillion parameter model for example. That could be very difficult to replicate even though ever
75.
by oofbey 5mo ago
But how can these poor googlers be expected to sift through the thousands of research papers published on these topics to find relevant citations? They don’t have time for such trivialities. They have far more important work to be doing no
76.
by oofbey 5mo ago
> It doesn’t decide to do something and then do it, it just outputs text. We can debate philosophy and theory of mind (I’d rather not) but any reasonable coding agent totally DOES consider what it’s going to do before acting. Reasoning.
77.
by oofbey 6mo ago
Nobody does it because it’s expensive. If you remove the requirement for perfect reproducibility you open the door to lots of optimizations. Most people prefer faster cheaper results over perfect reproducibility. When the model is intrinsic
78.
by oofbey 6mo ago
They didn’t say “your experience is not worse” but they did frequently say “just turn reasoning effort back up and it will be fine”. And that pretty explicitly invalidates all the (correct) feedback which said it’s not just reasoning effort
79.
by oofbey 6mo ago
Pickups are a fine place to start. If they’re successful they’ll add other kinds of cars over time. Building a whole new car company is extremely risky. Picking the first model correctly is extremely important. I hope they got it right. My
80.
by oofbey 6mo ago
There’s certainly some benefit in a declarative language for creating charts from SQL. Obviously this doesn’t do anything that you can’t also do easily in R or Python / matplotlib using about the same number of lines of code. But safel
81.
by oofbey 6mo ago
He’s kinda wrong. The facts he states are correct. Others are clearly exaggerating or making mistakes. But he jumps from that to the incorrect conclusion that “data centers don’t use water”. There are water accounting games on how much wate
82.
by oofbey 6mo ago
The thing Benn totally avoids is ever putting decibel levels on anything he measures. Yes NASA showed clearly that infrasound is dangerous at 140 dB (omg of course) and at 100dB it’s annoying and has psychological effects. But all the rese
83.
by oofbey 6mo ago
What would be a good use of AI? Writing code to do the modeling?
84.
Axios vulnerability with CVSS 10 over stated?(aikido.dev)
1 points by oofbey 6mo ago|1 comments
85.
by oofbey 6mo ago
CVSS of 10 sounds like this is another log4j, but it seems like the exploit requires some other unusual circumstances to work
86.
by oofbey 6mo ago
They absolutely IMPLY it’s becoming self aware, while not stating it explicitly. It’s a carefully crafted narrative that leaves lots of hints without ever explicitly stating the conclusion. Section 4.4.2: “we find this overall pattern of be
87.
by oofbey 6mo ago
This. One hundred percent this. It’s a carefully crafted story trying to convince people it’s showing signs of becoming self aware. But it’s a complete crock. They ran 1,000 tests where they left the answer key out for the agent to find, an
88.
by oofbey 6mo ago
My guess is it’s just incompetence. Imagine you’re in charge of ROCm and your boss asks you how it’s going. Do you say good things about your team and progress? Do you highlight the successes and say how you can do all the major things CUDA
89.
by oofbey 6mo ago
Years. They neglected ROCm for soooo long. I have friends who worked there 5+ years ago who tried desperately to convince execs to invest more in ROCm and failed. You had to have your head stuck pretty deep in the sand back then to not see
90.
by oofbey 6mo ago
Anthropic marketing is working very well. They are strongly incentivized to say their model is too powerful to release even if it’s not. It’s almost standard practice these days.

More ›