wonnage

5 ms·

31.
by wonnage 1mo ago
It turns out what’s fine for companies to do with individuals at relatively small scales is not fine for companies to do at massive scale and leverage
32.
by wonnage 1mo ago
Doing this may give you understanding but won’t really teach you how to understand. For students (as the article is focusing on them), the latter is equally important.
33.
by wonnage 1mo ago
More likely they know about but cannot fix it without tanking performance. It also effectively pigeonholes them into coding at the precise time they’re trying desperately to expand into general office work. You simply cannot use opus to gen
34.
by wonnage 1mo ago
they meant application performance
35.
by wonnage 1mo ago
Didn’t Kimi3 release a week before opus 5?
36.
by wonnage 1mo ago
PUA is short for pick up artist but has expanded to cover anyone using negging to convince you into doing something you didn’t want
37.
by wonnage 1mo ago
“asking the right questions” is built on years of experience doing the things now being offloaded to AI “asking the right questions” is also a moving target with each model release People simply underestimate the value of doing the work and
38.
by wonnage 2mo ago
Software stopped improving in 2016 and now AI is here to accelerate the downfall
39.
by wonnage 2mo ago
This is actually completely useless slop and any illusion of usefulness is misguided
40.
by wonnage 2mo ago
The package cache proxy is usually used to fetch from the public repository (npm, rubygems, etc.) so I think it could be feasible to craft some package metadata to trick it into GETing unexpected things. PUT/POST could be possible via
41.
by wonnage 2mo ago
https://arstechnica.com/ai/2025/06/anthropic-destroyed-milli... The scanning process destroys the book.
42.
by wonnage 2mo ago
Me when I construct my own strawman to avoid the topic Nowhere in GP comment was funding even mentioned
43.
by wonnage 2mo ago
But of course they’re excused for this little boo-boo whereas if kimi were caught doing the same thing it’d be an international incident
44.
by wonnage 2mo ago
So it’s even better, they’re 10x levering the money! Clearly I wasn’t thinking big enough
45.
by wonnage 2mo ago
Looks like you discovered an infinite money glitch! As long as you’re selling things at a profit, all you need to do is take those profits and give them to your customers to buy more things, repeat the loop a few times and you can become a
46.
by wonnage 2mo ago
The CI/CD costs are an interesting take, most of the rebuttals to AI ROI are something like “more code doesn’t mean more value!” but if you are charging per CI run then it actually does! Particularly with CI and extra testing being the
47.
by wonnage 2mo ago
It applies to software engineering too in things like autoscaling. I’ve seen many DDOS outages caused not by the attack itself but the wave of autoscaling that takes things down (often well after the actual attack ended).
48.
by wonnage 2mo ago
The verbosity itself seems to be a problem with post-4.6 Claude rather than a general issue with all LLMs. and IME the yak shaving is due to an overly generic prompt. We have a generic automated review bot but it’s prompted to only look for
49.
by wonnage 2mo ago
Interestingly if you look at the exploitgym repo ( https://github.com/sunblaze-ucb/exploitgym/blob/e5ea7c233a4d... ) the intended run mechanism is orchestrated by some python scripts which run agents against va
50.
by wonnage 2mo ago
Didn’t they explicitly remove alignment guardrails for this test? From the press release: > These deployment safeguards were intentionally not enabled during this evaluation because it was aimed at testing cyber vulnerabilities
51.
by wonnage 2mo ago
It’s one thing to predict a crash, it’s another to actually make money off of it. If you’ve watched The Big Short even those who predicted the housing crisis correctly nearly lost their shirt before and after it happened because timing the
52.
by wonnage 2mo ago
It’s probably not even a good idea to try and defend against the bubble by switching up your stock allocation. After all the whole reason passive investing works is that active investment rarely beats the market and if you’ve just been in S
53.
by wonnage 2mo ago
Reddit post so take with a grain of salt but this does seem possible. But unclear whether this is actually a viable architecture https://www.reddit.com/r/LocalLLaMA/comments/1t8s83r/nvidia_...
54.
by wonnage 2mo ago
I mean the questions are on random movie trivia, the alternative is just guessing. I think you're overthinking it.
55.
by wonnage 2mo ago
AI cannot tell you why it did something. If you ask it why it didn't do X even though your prompt said so, it can come up with some plausible sounding explanation, which might even be correct, but there's a fundamental impedance m
56.
by wonnage 2mo ago
Not sure if this is still up to date (2023), but https://arxiv.org/abs/2307.03172 shows that performance degrades mostly in the middle of the context. Anecdotally I've been stuck in that situation of being at 400-
57.
by wonnage 2mo ago
Agree that the study design is flawed. Going with the possibly-hallucinated AI answer is rational as long as you know the hallucination rate isn't 100% (or whatever % you get after factoring in the monetary rewards they introduced late
58.
by wonnage 2mo ago
They provide a sample of hallucinated answers from ChatGPT at the end of the study.
59.
by wonnage 2mo ago
This is a useless post-hoc rationalization. "It worked out, so it doesn't matter". You're trying to galaxy brain yourself into ignoring the obvious conclusion. The point is that if you were starting a new TUI LLM harness
60.
by wonnage 2mo ago
The top marginal tax rate in the US was 90% for most of the post WWII 20th century and that didn’t seem to hurt anyone. Invented transistors, went to the moon, built interstate highway system, mass construction of nuclear power, and became

More ›