milkshakes

7 ms·

1.
Unless there is a coordinated slowdown, human extinction seems likely(x.com)
2 points by milkshakes 6d ago|0 comments
2.
by milkshakes 8d ago
see https://en.wikipedia.org/wiki/XKeyscore for the sota from 20 years ago
3.
by milkshakes 8d ago
if your threat model includes the NSA i don't think your choice of CDN is going to make a difference
4.
by milkshakes 9d ago
i predict this perspective will not age well at all
5.
by milkshakes 11d ago
i am a big fan of llms and the possibilities they enable. but i also find this type of behavior extremely rude! ai;dr for life. :is-your-human-around: is my preferred emoji for reacting to such behavior
6.
by milkshakes 11d ago
i'm not blaming all problems here, i'm blaming the allocation of surplus. it does seem at least at surface level that this is driven by capitalism?
7.
by milkshakes 11d ago
> The problem is not that AI is evil, but most of the people would have to learn new skills and adapt because the profession or job they trained at college doesn't exists anymore. isn't the problem then capitalism? the promise
8.
by milkshakes 11d ago
> data centers are purely extractive please, tell me more > people in the community, or at least in the surrounding area, will play golf there going to go out on a limb here and guess that net-net, a datacenter provides more benefits
9.
by milkshakes 19d ago
brace yourself
10.
by milkshakes 26d ago
which of these do you represent?
11.
by milkshakes 26d ago
if a teenager did this, the police would show up, but in the end the feds or ic would intervene and recruit him
12.
by milkshakes 26d ago
justice for who exactly?
13.
by milkshakes 1mo ago
lol
14.
by milkshakes 1mo ago
inference is the cheap part; training is expensive. what compute infrastructure would train this mythical magic model?
15.
by milkshakes 1mo ago
see: https://en.wikipedia.org/wiki/Pharmacy_benefit_management
16.
by milkshakes 1mo ago
lately, i like to try to spend the morning talking through and framing up a few very detailed initiatives for parallel efforts that each spawn and orchestrate their own tasks, then feed them to /goals and go for a nice walk in the park
17.
by milkshakes 1mo ago
you can't downvote a submission it's just that nobody upvoted it
18.
by milkshakes 1mo ago
> models are still relatively easy to monitor and constrain are they?
19.
The OpenAI–Hugging Face Incident [video](youtube.com)
90 points by milkshakes 1mo ago|22 comments
20.
Ten advances in mathematics and theoretical computer science(openai.com)
632 points by milkshakes 1mo ago|942 comments
21.
Ten advances in mathematics and theoretical computer science(openai.com)
464 points by milkshakes 2mo ago|0 comments
22.
by milkshakes 2mo ago
they also have a pretty handy button right next to the error to ftfy
23.
by milkshakes 2mo ago
i guess what i'm curious about is where you would draw the line, and why. how do you define a user agent? is it desirable or not for users to be able to discover and access resources and communities on the internet using the tools, for
24.
by milkshakes 2mo ago
i guess the thing i'm most confused about is what is the higher level goal here. 1 in 8 humans on the planet are experiencing the "web" through chatgpt alone. many have migrated to purely agentic workflows. is the goal for yo
25.
by milkshakes 2mo ago
start, not finish
26.
by milkshakes 2mo ago
http://www.incompleteideas.net/IncIdeas/BitterLesson.html > One thing that should be learned from the bitter lesson is the great power of general purpose methods, of methods that continue to scale with increased com
27.
by milkshakes 2mo ago
start here: https://huggingface.co/blog/mlabonne/abliteration
28.
by milkshakes 2mo ago
what top comment?
29.
by milkshakes 2mo ago
nyt is no fan of AI, but they cover this: https://www.nytimes.com/2026/07/09/business/china-russia-ai-...
30.
by milkshakes 2mo ago
this is quite literally reward hacking. the model, under evaluation with cyber capabilities enabled, used those capabilities to simply bypass the exercise entirely and aim straight for the source of the flag. the CTF equivalent back in the

More ›