wrs

6 ms·

91.
by wrs 3mo ago
What's even worse than so many students cheating with AI is that I suspect a substantial portion of them don't even think that's "cheating".
92.
by wrs 3mo ago
I really don't think this project was meant to be taken seriously.
93.
by wrs 3mo ago
No relation. GUID is just a format for a 128-bit unique number, used throughout the software industry. This is a specific 64-bit number assigned to your Windows device.
94.
by wrs 3mo ago
I suspect when those 10-digit wire transfers start arriving in your bank account, your attitude changes rapidly.
95.
by wrs 3mo ago
>Comments should be passed to the model with clear role boundaries that prevent them from being interpreted as system-level directives. Well, such clear boundaries would solve lots of problems. But those don’t exist, do they?
96.
by wrs 3mo ago
I thought the point was not that Mythos finds more vulnerabilities, but that it can exploit them much more successfully. I thought the report showed it didn’t find much more than Opus 4.8. (Or did I misread?)
97.
by wrs 3mo ago
My current technique, which seems to have improved maintainability, is to guide Claude to write commit messages specifically focused on "why this was done", "what changed in the theory of operation", and "what chang
98.
by wrs 3mo ago
Replace science with software development and this just reads like half of HN right now.
99.
by wrs 3mo ago
I often tell Claude Code to look at previous sessions in ~/.claude and it’s happy to jq/grep its way through them with no special tool. But being more efficient is always good.
100.
by wrs 3mo ago
Apparently, though not very carefully. The "particularly large LLM generated code churn" in the ram library, for example, is the LLM being used to simply git-revert a change that was not originally done by an LLM.
101.
by wrs 3mo ago
The manifold is in the middle (“small input space is expanded onto a big manifold and contracted again”) so f(manifold) would need to be in the middle too.
102.
by wrs 3mo ago
Obviously your distro isn’t using cryptsetup-luksSuspend.
103.
by wrs 3mo ago
I'm honestly surprised they didn't put a flat-earther in charge of NASA.
104.
by wrs 3mo ago
I think that’s what happens if you’re on Cloudflare DNS, due to a nerdy dispute they’re having with archive.is.
105.
by wrs 3mo ago
Well, not once the scores become meaningless because everyone assumes they cheated.
106.
by wrs 4mo ago
Anthropic submitted a long, thoughtful framework proposal, which everyone seems to be ignoring in favor of hot takes like “they asked for this!” No they didn’t, not at all.
107.
by wrs 4mo ago
So…you can transform it into a threaded interpreter? This seems like a ton of LLM verbiage making two simple and very standard techniques sound artificially profound.
108.
by wrs 4mo ago
>"The decisions were based upon thorough risk assessments..." Both decisions? Or just the walkback one?
109.
by wrs 4mo ago
They're also anti-"freedom". (Individual freedom being one of the paramount values of the military.) (edit: Sorry, thought the sarcasm would be obvious…)
110.
by wrs 4mo ago
This is a way of using LLMs that doesn’t seem appreciated enough: prototyping a refactor or other new approach to see how it goes. Even if you then rewrite it from scratch, it’s a big timesaver because you’re using the new knowledge to do t
111.
by wrs 4mo ago
Agreed. What happened to everything else on comms failure?
112.
by wrs 4mo ago
Perhaps, but there are optics only when someone sees. It's not unusual to fund retention increases using some of the budget freed up by a layoff, but that won't be explicitly stated in a public announcement.
113.
by wrs 4mo ago
How does that follow? Plenty of open source software runs in a commodity datacenter. This is about the API bottleneck, not the physical location of the GPU.
114.
by wrs 4mo ago
Especially considering the millions of 2026-class data center GPUs that massively overinvested companies are currently buying, which will be obsolete in a few years.
115.
by wrs 4mo ago
That seems like a fail-safe interpretation of communication: if there is no communication, stop the train. But that's a special case. GSM-R is much more than line authorities. >GSM-R is a secure platform for voice and data communi
116.
by wrs 4mo ago
>This special mobile communication standard is designed to make communication fail-safe Mmm, nope.
117.
by wrs 4mo ago
IIRC from the Anthropic report, the alleged danger of Mythos isn’t that it finds more vulnerabilities than previous models, but that it’s significantly more successful at exploiting them. Which this doesn’t seem to test.
118.
by wrs 4mo ago
The big indexes have a cooldown period before buying, which these scam artists attempted to radically shorten before the IPO, but they failed (at least partially). The S&P500, for example, will delay a year before adding it to the index
119.
by wrs 4mo ago
Right, thus it is actually “open source”, and they shouldn’t need to invent a new term “open training source”. But the others have already effectively trashed the meaning of “open source”.
120.
by wrs 4mo ago
Well, 2-4 hours of autonomous work is outside my comfort zone to start with. But have you tried Claude Code’s “auto mode”? I haven’t seen this premature stopping since it was introduced.

More ›