frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

California's AB 412 Still Demands Developers Do the Impossible

https://www.eff.org/deeplinks/2026/06/californias-ab-412-still-demands-developers-do-impossible
6•hn_acker•1h ago

Comments

yladiz•8m ago
> California lawmakers are again considering A.B. 412, a bill that would require AI developers to identify and disclose copyrighted works used to train generative AI systems.

> The problem this year is the same as last year: it’s practically impossible to comply with this law. The bill demands information that often does not exist, and cannot realistically be obtained.

> Its definition of “developer” extends to anyone who makes a generative AI model available to Californians.

I get that this would burden up-and-coming companies that want to train new models, but in general I don't think it's a bad thing that a company needs to know where the material they train their model comes from, and know its copyright status, and if it's actually an impossible problem then maybe the whole system is unworkable. Assuming that model training isn't fundamentally considered fair use, how else can you approach this problem?

ElevenLathe•1m ago
It's wild how software BoM is taking off at the same time that LLM BoM is being declared literally impossible. IMO the threat model is roughly the same: if you can't account for the provenance of all the text in your training set, how can say that it hasn't been poisoned?

Configuration flags are where software goes to rot

https://00f.net/2026/04/11/config-flags/
1•birdculture•23s ago•0 comments

CrankGPT: A human-powered local and private AI solution

https://crankgpt.com/
2•_ihaque•1m ago•0 comments

Show HN: Group of Death – WorldCup bracket predictor with shareable URLs

https://groupofdeath.gg/
2•sparc24•1m ago•0 comments

FeOS: A 16-Bit OS Written in ASM (and a custom language)

https://github.com/anhumandev/feos
1•anhuman•2m ago•1 comments

Neural-somatic simulations of history's greatest minds

https://www.opendria.com/
1•opendria•3m ago•0 comments

Understanding why autism symptoms sometimes improve amid fever

https://news.mit.edu/2024/understanding-why-autism-symptoms-sometimes-improve-amid-fever-0523
2•thunderbong•3m ago•0 comments

</> Htmx – Code is Cheap(er)

https://htmx.org/essays/code-is-cheap/
2•leephillips•6m ago•0 comments

The circus freaks of open source

https://drewdevault.com/blog/Circus-freaks-of-FOSS/
2•Tomte•7m ago•0 comments

The round-the-world escape from Pearl Harbor

https://signoregalilei.com/2026/05/17/the-round-the-world-escape-from-pearl-harbor/
1•surprisetalk•7m ago•0 comments

SignalScout

https://signalscout.be
2•AIGENIZE•8m ago•0 comments

My thoughts on the "focusgroup" attribute proposal

https://www.stevefrenzel.dev/posts/my-thoughts-on-the-focusgroup-attribute-proposal/
1•speckx•12m ago•0 comments

A curated list of AI for developers

https://github.com/ai-for-developers/awesome-ai-coding-tools
2•dariubs•13m ago•0 comments

Stop Using Conventional Commits

https://sumnerevans.com/posts/software-engineering/stop-using-conventional-commits/
5•jsve•13m ago•0 comments

Federal audit reveals NIST's NVD is plagued by poor planning and duplication

https://cyberscoop.com/nist-nvd-audit-mismanagement-duplication/
2•u1hcw9nx•14m ago•0 comments

AI investment 2nd round, from GPU to power·industrial goods·space

https://www.ooooo.law/board/9?lang=en
1•haebom•16m ago•0 comments

What Dot.com Bandwidth Taught Me About the AI Token Cost Panic

https://www.axamy.com/blog/bandwidth-tokens
1•jhonovich•17m ago•1 comments

MCemm a GEMM (General Matrix Multiply) Kernel Generato

1•Okerew•17m ago•0 comments

Clean City Law: Secrets of SãO Paulo Uncovered by Outdoor Advertising Ban

https://99percentinvisible.org/article/clean-city-law-secrets-sao-paulo-uncovered-outdoor-adverti...
1•chistev•18m ago•0 comments

Agentic communication protocol – why A2A sucks

https://asimovaddendum.substack.com/p/agents-need-a-public-square
1•ilan_s•19m ago•1 comments

The cognitive benefit of a window view

https://www.sciencedirect.com/science/article/pii/S0167268125004755
1•PaulHoule•19m ago•0 comments

Leak Reveals Microsoft Wants Its AI to Be 'Addictive'

https://kotaku.com/microsoft-ai-scout-addictive-satya-nadella-404-media-copilot-2000702924
7•thm•20m ago•0 comments

Justices validate SEC's use of disgorgement in securities enforcement

https://www.scotusblog.com/2026/06/justices-validate-secs-use-of-disgorgement-in-securities-enfor...
3•jawns•20m ago•0 comments

Russian Satellites Have Been Jamming GPS Signals Across Europe, Scientists Say

https://www.nytimes.com/2026/06/05/world/europe/russia-satellites-gps-interference-europe.html
5•ripe•20m ago•1 comments

Show HN: 2Draw – co-operative pictionary built with TLDraw inspired by Drawful

https://2draw.ritzademo.com/
1•sixhobbits•20m ago•0 comments

An Unbiased OSS Benchmark for Code Review Agents

https://codereview.withmartian.com/
1•forks•21m ago•0 comments

Show HN: Lich, start a dev stack per coding agent in parallel

https://github.com/RPate97/lich
1•patethegreat•22m ago•0 comments

When Does Federal Debt Reach Unsustainable Levels? Spring 2026 – Onward

https://budgetmodel.wharton.upenn.edu/p/2026-06-02-when-does-federal-debt-reach-unsustainable-lev...
1•fzliu•22m ago•0 comments

Elevated errors on many Claude models

https://status.claude.com/incidents/fprlnsvdnr2k
5•corvad•22m ago•0 comments

Show HN: Agents Remember – Git-aware memory for coding agents

https://github.com/Foxfire1st/agents-remember-md
1•foxfire_1st•23m ago•0 comments

Celebrity Border Collie with 1.5M followers dognapped and eaten

https://www.independent.co.uk/asia/china/chutou-dog-theft-china-border-collie-b2989675.html
2•SilverElfin•26m ago•1 comments