Hacker News: with

New comment by with in "AI is forcing us to pay off 40 years of technical debt"

with — Wed, 04 Nov 2026 18:07:49 +0000

and creating 40 years more

New comment by with in "Progress on provable inference for large language models"

with — Wed, 04 Nov 2026 17:44:15 +0000

isn't model substitution a vendor accountability thing? and token tampering, is that not infra security?

do either of these need 70-95x the GPU time in "zksnark" proofs?

New comment by with in "An MCP for Telling the Time?"

with — Mon, 07 Sep 2026 00:04:19 +0000

the MCP equivalent of is-even (https://www.npmjs.com/package/is-even) on NPM

New comment by with in "[dead]"

with — Sun, 06 Sep 2026 23:55:55 +0000

it's incredibly low effort and very high yield. it only keeps boosting their SEO. they have enough human traffic to where they can totally get away with it too

New comment by with in "App Review Shenanigans"

with — Sun, 06 Sep 2026 23:54:24 +0000

the only way I can describe the modern Apple App Review process is "silly". necessary, but very silly. I was rejected AND had my account terminated for having Firebase Remote Config in my app, and they thought I was hiding features from their review. when I sent evidence they immediately lifted the suspension.

It's just a hobby for me (just $100/month in IAP revenue rn) but I really don't understand their system at all.

New comment by with in "Pivot to AI safety, I beg you"

with — Sun, 06 Sep 2026 23:47:09 +0000

> We will need MILLIONS of people in the field.

it's a real issue but I'm not sure it's the mass mobilization problem the article makes it out to be. pretty sure it's growing extremely fast, and it's a new field in general. the models will keep growing in capability because of human game theory (I need to make my model better because my competitors are). it's not practical to expect millions in headcount to move from frontier capability research to safety

New comment by with in "Copilot edited an ad into my PR"

with — Mon, 30 Mar 2026 05:29:18 +0000

Everyone is doing this now. Granted, on Codex / Claude Code, you can disable it, it’s not the default to have it disabled. For some reason on Cursor, they keep shoving the “Made with Cursor” into my PR description despite me disabling attribution, which looks really stupid on a work PR.

I’m so tired of all this BS. Why did this become normal? and how do we not read this as cheap advertising?

New comment by with in "Ask HN: Why Isn't Everything Public?"

with — Mon, 30 Mar 2026 04:35:38 +0000

yes, but the same people who have the power to make those same things public are the ones benefiting from the same fraud, waste, and corruption

New comment by with in "Miasma: A tool to trap AI web scrapers in an endless poison pit"

with — Mon, 30 Mar 2026 04:26:05 +0000

> do these types of techniques really work?

They have been proven to: https://www.anthropic.com/research/small-samples-poison

New comment by with in "The Cognitive Dark Forest"

with — Mon, 30 Mar 2026 04:21:54 +0000

if the idea can just be obliterated by an LLM, there was never a moat to begin with

New comment by with in "AI isn't killing jobs, it's 'unbundling' them into lower-paid chunks"

with — Mon, 30 Mar 2026 04:15:46 +0000

Just further proof that context is the real moat, not intelligence. All the models are already converging to be equally intelligent and that will only continue. GPT 5.4 / Opus 4.6 are the first two models I’ve used where I’m like, yeah, with the right spec/context they can pretty much do anything.

The “bundle” or “context” is the value.

New comment by with in "A decade of Docker containers"

with — Sun, 08 Mar 2026 07:38:50 +0000

nice share!

New comment by with in "A decade of Docker containers"

with — Sun, 08 Mar 2026 05:50:16 +0000

docker is bloated. i'm almost certain half of every image is dead weight. unused apt packages, full distros for a single binary, shell configs nobody touches. but the incentive is to make things work, not make them small. so bloat wins.

still, i use it every day and i don't see what replaces it. every "docker killer" solves one problem while ignoring the 50 things docker does well enough.

New comment by with in "Warn about PyPy being unmaintained"

with — Sun, 08 Mar 2026 05:08:51 +0000

Thanks, I also saw this as PyPI and was confused, lol

New comment by with in "Training students to prove they're not robots is pushing them to use more AI"

with — Sat, 07 Mar 2026 23:01:11 +0000

Fair point, the "tuned to flag aggressively" claim was speculative on my part. Turnitin's own documentation says they favor false negatives over false positives.

That said, their accuracy claims have been disputed before. Inside Higher Ed [1] reported that Turnitin's real-world false positive rate was higher than originally asserted, and the company declined to disclose the updated number. And, USD also noted that while Turnitin claimed <1% false positives, a Washington Post investigation found a 50% rate on a smaller sample, and that non-native English speakers / neurodivergent students get flagged at higher rates [2].

Now, those are from 2023 and the product (and AI in general) has been updated drastically since. But the broader incentive problem holds even if the detector itself is conservatively tuned. The product is a black box. And the downstream cost of errors falls entirely on students, not on Turnitin's renewal rate. You don't need aggressive tuning for the incentive structure to be broken.

[1] https://www.insidehighered.com/news/quick-takes/2023/06/01/t...

[2] https://lawlibguides.sandiego.edu/c.php?g=1443311&p=10721367

New comment by with in "Training students to prove they're not robots is pushing them to use more AI"

with — Sat, 07 Mar 2026 21:30:01 +0000

nobody's asking who profits from false positives. these AI detection vendors have a direct financial incentive to flag aggressively. more flags = "more value" = more school contracts renewed. same playbook as selling antivirus to your grandma. sell fear, charge per seat, and make the false positive rate someone else's problem.

New comment by with in "When does MCP make sense vs CLI?"

with — Sun, 01 Mar 2026 23:59:17 +0000

MCP is a completely useless abstraction, and I’m not sure why anyone would push for it over basic cli tools / skills.

New comment by with in "Show HN: The Dot"

with — Sun, 01 Mar 2026 07:57:57 +0000

this looks chaotic. I love it

New comment by with in "Microgpt"

with — Sun, 01 Mar 2026 07:24:26 +0000

yep, agreed. wasn’t knocking the project at all, it’s great for exactly that purpose

New comment by with in "Microgpt"

with — Sun, 01 Mar 2026 07:09:16 +0000

"everything else is just efficiency" is a nice line but the efficiency is the hard part. the core of a search engine is also trivial, rank documents by relevance. google's moat was making it work at scale. same applies here.