frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

The JTAG in your Qualcomm/Snapdragon device's USB port

https://www.linaro.org/blog/hidden-jtag-qualcomm-snapdragon-usb/
1•denysvitali•50s ago•0 comments

Mechanistic Interpretability of Emotion Inference in Large Language Models

https://arxiv.org/abs/2502.05489
1•cainxinth•3m ago•0 comments

Why are AI agents not allowed to roam freely on the internet?

https://potent-menu-ea7.notion.site/Why-are-AI-agents-not-allowed-to-roam-freely-on-the-internet-like-humans-222903aa31ef8053a1b9dc5da51e0770
1•orliesaurus•3m ago•0 comments

Datadog's $65M/year customer mystery solved

https://blog.pragmaticengineer.com/datadog-65m-year-customer-mystery/
1•thunderbong•4m ago•0 comments

OpenTelemetry Is Great, but Who the Hell Is Going to Pay for It?

https://www.adatosystems.com/2025/02/10/who-the-hell-is-going-to-pay-for-this/
2•thunderbong•7m ago•0 comments

How Long Contexts Fail

https://www.dbreunig.com/2025/06/22/how-contexts-fail-and-how-to-fix-them.html
1•ulrischa•8m ago•0 comments

Brain rot isn't new – but now we're all talking about it

https://www.theguardian.com/books/2025/jun/30/brain-rot-isnt-new-but-now-were-all-talking-about-it
1•Bluestein•9m ago•0 comments

AI Economy and Mass Unemployment

https://claude.ai/share/dbefc4de-5ba3-4f0a-9e38-e33e2a15903b
2•VikRubenfeld•11m ago•0 comments

Meta Joins Kotlin Foundation

https://engineering.fb.com/2025/06/30/android/meta-joins-kotlin-foundation/
1•bundie•14m ago•0 comments

Ask HN: How have you shared computers with your young child (~3 to 5)

2•msencenb•14m ago•0 comments

Don't Be Ashamed to Say "I Don't Know"

https://www.thecoder.cafe/p/i-dont-know
1•gpi•17m ago•0 comments

I'm a software engineer – what next?

https://whatnext.dev/
2•mooreds•19m ago•0 comments

Rubik's Cube Solver

https://rubiks-cube-solver.com/
2•austinallegro•21m ago•0 comments

Show HN: Validated Daily SaaS Ideas

https://www.dailysaasideas.com/
2•sandandcode•23m ago•0 comments

What's the difference between bio and non bio detergent? (2022)

https://smol.com/uk/stories/whats-the-difference-between-bio-vs-non-bio-detergent
2•mooreds•23m ago•0 comments

Amateurs Talk Algorithms, Professionals Talk Data Cleaning

https://progressandpoverty.substack.com/p/amateurs-talk-algorithms-professionals
3•hliyan•24m ago•0 comments

Show HN: BugBlaze – a CLI tool that explains your code errors using AI

https://github.com/MatthewTheCoder1218/bugblaze
1•lilprince1218•24m ago•0 comments

Ask HN: What's the 2025 stack for a self-hosted photo library with local AI?

9•jamesxv7•24m ago•6 comments

The Bear Manifesto – Herman's Blog

https://herman.bearblog.dev/manifesto/
1•FigurativeVoid•24m ago•0 comments

Alex Trebek Stamps, Sheet of 20

https://store.usps.com/store/product/alex-trebek-stamps-S_485304
1•mooreds•25m ago•1 comments

Show HN: Kichan.ai a free Chrome extension to generate and inject scripts

https://kichan.ai/
1•k1rd•26m ago•0 comments

Show HN: Attach Gateway – one-command OIDC/DID auth for local LLMs

https://github.com/attach-dev/attach-gateway
2•hammadtariq•26m ago•0 comments

Zig Community Mirrors

https://ziglang.org/download/community-mirrors/
2•todsacerdoti•27m ago•0 comments

The Simple Math Problem We Still Can't Solve

https://www.quantamagazine.org/why-mathematicians-still-cant-solve-the-collatz-conjecture-20200922/
2•chirau•30m ago•1 comments

They don't make 'em like that any more: Sony DTC-700 audio DAT player/recorder

https://kevinboone.me/dtc-700.html
5•naves•31m ago•0 comments

User-friendly and privacy-friendly LLM experience?

https://tildes.net/~comp/1orz/user_friendly_and_privacy_friendly_llm_experience
1•PaulHoule•31m ago•0 comments

Ask HN: Stock Android tablet free of bloatware?

3•miki_tyler•33m ago•1 comments

Alice's Adventures in a Differentiable Wonderland

https://arxiv.org/abs/2404.17625
1•henning•33m ago•0 comments

Why We Should Care About This War over the Future of Money

https://gizmodo.com/why-you-should-care-about-this-war-over-the-future-of-money-2000622009
2•rntn•34m ago•0 comments

Lua 5.5.0 (Beta) Released

https://www.lua.org/work/#5.5.0
2•dottrap•35m ago•1 comments
Open in hackernews

Congress might block state AI laws for a decade

https://techcrunch.com/2025/06/27/congress-might-block-state-ai-laws-for-a-decade-heres-what-it-means/
22•wslh•4h ago

Comments

bgwalter•3h ago
The "AI" czar has been funding telehealth companies and is now probably complicit in the largest takeaway of states' rights ever:

https://www.wsj.com/us-news/law/executives-from-adhd-startup...

https://sfstandard.com/2024/06/13/telemedicine-adderall-vyva...

He seems to have a history of investing in companies that get into trouble:

https://techcrunch.com/2016/09/13/ceo-david-sacks-on-moving-...

It is quite ironic that Republicans have been complaining about the erosion of states' rights forever. I suppose that now a telehealth investor decides the future of "AI" safety and copyright laundering.

kurthr•3h ago
Yeah, don't complicate it, it's just corruption.
RajT88•3h ago
They do not care about all state's rights, only certain ones. You know the ones.
meager_wikis•3h ago
The AI czar is symbolic of the terrible year it's been for American capitalism. Fealty, corruption, regulatory capture. Extracting value from the less powerful instead of creating it.
schmidtleonard•2h ago
It was an excellent year for American Capitalism. Get outa here with that "capitalism fights for the little guy" rubbish. We all love a good disruption story, we all love to personally win the game, but on a systems level it is crystal clear that on net the incentives overwhelmingly point in the opposite direction.

The objective function of capitalism is weighted by wealth. "Capitalism is about doing what other people want" is the wealth_gini=0 limit. "Capitalism is about doing what rich people want" is the wealth_gini=1 limit. The USA has a wealth_gini of about 0.85. What do rich people want that poor people don't? Stuff their pockets by extracting value from the poors, of course, which is exactly what we see here. Canceling redistributive programs and taxes is about letting the market do more of what it wants to do: serve the rich, screw the poor.

meager_wikis•2h ago
Would you consider it to be a good year when it leaves millions believing capitalism does not better their lives? We are agreeing.
DangitBobby•2h ago
You see, it's a small government move in this case because only states they don't like are bothering to attempt AI regulations.
Fin_Code•3h ago
If this was left at the state level some elected official would say the devil is in the code and we need to ban it.
Analemma_•3h ago
Which state official has said "the devil is in the code and we need to ban it", can you name them?
paulvnickerson•3h ago
This makes sense from a national security perspective. China is a unified AI strategy, and America can either lead or fall behind, in which case we will all be using Chinese models that bend reality towards CCP's preferred idealogy (look into Deepseek censorship to see how this will work). If AI companies have to navigate 50 different AI regulatory regimes it will bog down progress.

I know my opinions are often unpopular here on HN. It's unfortunate that so many people resort to flagging them based on disagreements.

bgwalter•3h ago
I don't buy that. You can block Chinese companies. Human influencers are more effective than "AI" slop, etc.

And again (you didn't make that argument, just for completeness' sake): Military applications aren't affected anyway. They can train on whatever they want as long as the products stay in the military and do not compete with writers and artists etc.

silverquiet•3h ago
I actually tend to agree with this sentiment, but the fate of the Tiktok ban should probably be instructive here. America (at least the current government) is not really interested in world or technological leadership anymore.
blibble•2h ago
> in which case we will all be using Chinese models that bend reality towards CCP's preferred idealogy

quite how this is worse than what zuckerberg or elon want I really don't know

kelseyfrog•2h ago
What a tough nut to crack because almost every issue can be reduced to national security. What scissor do you use when deciding whether the national security perspective justifies other consequences like loss of freedom?
throwworhtthrow•1h ago
AI companies already need to navigate multiple regulatory regimes. Chinese companies are not exempt.

If a few states adopt more stringent rules, US-based companies will likely have an easier time meeting them than foreign companies. Don't you think Facebook and Google helped shape the CCPA to make compliance easier for them, perhaps to the detriment of foreign competition?

derekp7•3h ago
Wouldn't the interstate commerce clause automatically override any state level regulations on AI, except in cases where a company and its customers are all in-state?
tristan957•1h ago
Not too familiar with the law, but Southwest Airlines only operated within Texas, and existed without federal regulation for some time, but eventually the federal government regulated them too.
Ancalagon•3h ago
If Republicans were actually for states rights like they say when it comes to abortions then they should put their money where their mouths are and let states decide for themselves on AI.

Are yall in the comments agreeing with this actually astroturfing? I can smell the hypocrisy.

ChrisArchitect•1h ago
[dupe] https://news.ycombinator.com/item?id=44406186
tristan957•1h ago
Texas already did this at the municipal level[0]. Government power continues to be centralized whether it is Republicans or Democrats. Republicans seem to be the only ones marketing themselves as the party of small government though, which is a complete misnomer.

[0]: https://www.dlapiper.com/en/insights/publications/2025/06/te...