russfink

7 ms·

31.
by russfink 8mo ago
Okay, but what are the cybersecurity implications of this discovery? A new living-off-the-land approach? Resource exhaustion? Covert servers?
32.
by russfink 8mo ago
Sorry for the novice question, but how do you determine this is AI?
33.
by russfink 8mo ago
Not sure I understand the problem this is trying to solve. I can hash a 2Gb file with an empty hash placeholder, sign the hash and store it in the file. Then on decryption, zero out the placeholder, hash the whole file, and compare the res
34.
by russfink 8mo ago
Hug of death via Tom’s Hardware
35.
Log into 28 vintage computer systems in the browser / Interim Computer Museum(tomshardware.com)
3 points by russfink 8mo ago|2 comments
36.
by russfink 8mo ago
I must be the only one who thinks this, but this is the age of getting things done. I don’t have to worry about syntax or off by one errors, I tell it what to do and it generally does it, instantly! Oh, and I’m 57 and was programming the Co
37.
by russfink 8mo ago
sedutil-cli —yesIwantToEraseALLmydata $PSID /dev/sda1 or something like that.
38.
by russfink 9mo ago
This. Most people defer the solving of hard problems to when they write the code. This is wrong, and too late to be effective. In one way, using agents to write code forces the thinking to occur closer to the right level - not at the code l
39.
by russfink 9mo ago
zOMG dying laughing here
40.
by russfink 9mo ago
> With four treatment sessions spaced fortnightly, This is a clearer statement of “Every two weeks” than “bimonthly” or “semimonthly.” Brilliant!
41.
by russfink 10mo ago
Mathematically speaking, “no child left behind” is equivalent to “no child out in front.”
42.
by russfink 10mo ago
Except what if you don’t really grok those ffmpeg flags and the LLM tells you something wrong - how will you know? Or more common, send you down a re-encode rabbit hole when you just needed a simple clipping off the end?
43.
by russfink 10mo ago
Ten years old laptop? Pretty sure it has a TPM 2.0 on it.
44.
by russfink 10mo ago
Practical question: when getting the AI to teach you something, eg how attention can be focused in LLMs, how do you know it’s teaching you correct theory? Can I use a metric of internal consistency, repeatedly querying it and other models
45.
by russfink 10mo ago
Conspicuously missing is a direct mention of AI tools. Is MIT, like others, side-stepping the use of AI by students to (help them) complete homework assignments and projects?
46.
by russfink 10mo ago
Link to the About page that clearly describes the effort and rationale. https://missing.csail.mit.edu/about/
47.
by russfink 10mo ago
Back in the day, wustl.edu was seen as a leader in computer applications. Sad now that it cannot just create its own systems to handle its tasks, especially with AI’s around to offer coding help. Imagine spending a fraction of this money an
48.
by russfink 10mo ago
One trick I have tried is asking the LLM to output a specification of the thing we are in the middle of building. A commenter above said humans struggle with writing good requirements - LLMs have trouble following good requirements - ALL of
49.
by russfink 10mo ago
Damn … Crucial P3 plus and P5 plus support Opal 2.0 full disk encryption. This leaves Samsung standing nearly alone in the consumer market, except for some smaller names.
50.
by russfink 11mo ago
If it matters, many of their issues were apparently closed without comment. Here is an issue asking to resolve this. It also lists many of the commits. https://gitlab.freedesktop.org/xorg/xserver/-/work_items
51.
by russfink 11mo ago
May be a repeat here, but best proof I saw was inscribe a square with sides of length c inside another square, but rotated such that the interior square’s corners intersect the outer square’s edges. The intersecting points divide the outer
52.
by russfink 11mo ago
“Build me a fishing pole” instead of “catch me a fish” is the best advice this article has (sans my fishing metaphor).
53.
by russfink 11mo ago
Try the Developers remix. Total ear worm. https://youtu.be/ug4c2mqlE_0?si=qtqu7tOC7Xpw67aN
54.
by russfink 11mo ago
Is this a DoS risk - code that sends your build chain into an infinite loop?
55.
by russfink 11mo ago
I use LLMs to help navigate FFmpeg. It hasn’t failed me yet.
56.
by russfink 1y ago
Nowadays, do pass 1 (abstract, intro, headings, conclusions), upload the paper to your LLM, then ask questions interactively. LLMs can make step 3 a lot more efficient.
57.
by russfink 1y ago
I get the impression that somehow an attacker is able to inject this prompt (maybe in front of the actual coder’s prompt) in such a way to produce actual production code. I’m waiting to hear how this can happen - cross site attacks on the d
58.
by russfink 1y ago
“Brewing…” et tu, smart coffee maker?
59.
by russfink 1y ago
Ransom? Or is that too risky?
60.
by russfink 1y ago
Or stolen for ransom. That might fetch the largest payout of all.

More ›