frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

A Few Details Apple Didn't Mention During Its "Awe-Dropping" Event

https://sixcolors.com/post/2025/09/a-few-details-apple-didnt-mention-during-its-awe-dropping-event/
2•Bogdanp•4m ago•0 comments

AI agents are coming for your privacy

https://www.economist.com/by-invitation/2025/09/09/ai-agents-are-coming-for-your-privacy-warns-me...
1•b_e_n_t_o_n•6m ago•0 comments

Lawyer named Mark Zuckerberg sues Meta after repeated account shutdowns

https://nypost.com/2025/09/03/us-news/lawyer-named-mark-zuckerberg-sues-meta-over-claims-hes-impe...
3•1vuio0pswjnm7•16m ago•0 comments

Show HN: Claude-Powered Survival Analysis with One-Click Models (Onco-Insight)

https://dataize.me
1•DATAIZE•18m ago•0 comments

Tell HN: Avoid ostentatious eloquence and crisp structure or be slop

1•aniijbod•21m ago•0 comments

Greased Turkey, and Other Linux Kernel Release Names

https://handwiki.org/wiki/List_of_Linux_kernel_names
1•e-dant•22m ago•0 comments

LavaMoat – tools for sandboxing your dependency graph

https://github.com/LavaMoat/LavaMoat
1•TheFreim•26m ago•0 comments

Lessons from NPM's Security Failures

https://oneuptime.com/blog/post/2025-09-09-lessons-from-npm-security-failures/view
2•thunderbong•28m ago•0 comments

Ray nayler the mountain in the sea (scifi 2023)

https://www.raynayler.net/the-mountain-in-the-sea.html
2•coderatlarge•29m ago•2 comments

Gear Generator 2 Beta

https://geargenerator.com/beta/
1•jumpocelot•30m ago•0 comments

Show HN: GTA AI – Convert Your Photos to GTA Style Artwork

https://gtaai.app/
3•tomstig•31m ago•0 comments

Arcane 1.0 Released

https://github.com/ofkm/arcane
1•nichos•32m ago•0 comments

Show HN: Sayathing – Open-Source platform that gives your text a voice

https://github.com/kanthorlabs/sayathing
1•buglungtung•35m ago•0 comments

We taught AI a new modality – UI

https://computerhumaninteraction.com
1•jojule•36m ago•1 comments

The Jeopardy game that changed everything still haunts the show 15 years later

https://slate.com/culture/2025/09/jeopardy-game-watson-questions-final-ken-jennings.html
2•voisin•36m ago•0 comments

Australia to force porn websites to check users' ages under new internet rules

https://www.crikey.com.au/2025/09/09/porn-websites-legally-check-user-age-australia-teen-social-m...
3•t0lo•43m ago•1 comments

SelfHosting "Easy" Solutions

https://docs.google.com/spreadsheets/d/e/2PACX-1vRId9P6-c-XzMZQyzG6ROlpV804w-VzD685fQZQ-GSpMl9Duq...
1•indigodaddy•44m ago•0 comments

Full MCP tools support in ChatGPT

https://twitter.com/openaidevs/status/1965581442370707861
2•mfiguiere•44m ago•0 comments

Uncovering The Genes That Let Our Ancestors Walk Upright

https://www.nytimes.com/2025/08/27/science/human-evolution-ilium-bipedal.html
1•bookofjoe•47m ago•1 comments

An inner-speech decoder reveals some mental privacy issues

https://arstechnica.com/science/2025/08/an-inner-speech-decoder-reveals-some-mental-privacy-issues/
1•PaulHoule•47m ago•0 comments

I Hate My Friend

https://www.wired.com/story/i-hate-my-ai-friend/
1•arbuge•48m ago•0 comments

Show HN: Subway Wrapped. Your NYC transit year in review

https://www.subwaywrapped.com/
1•faisal_software•48m ago•0 comments

R-Zero: Self-Evolving Reasoning LLM from Zero Data

https://arxiv.org/abs/2508.05004
2•lawrenceyan•49m ago•0 comments

VCF East: Unix: A History and a Memoir by Brian Kernighan [video]

https://www.youtube.com/watch?v=WEb_YL1K1Qg
1•LordNibbler•49m ago•0 comments

Hyundai Raid Shows Trump Can't Deport His Way to a Manufacturing Boom

https://reason.com/2025/09/09/hyundai-raid-shows-trump-cant-deport-his-way-to-a-manufacturing-boom/
7•SanjayMehta•53m ago•0 comments

Russian drone alert in Poland before Belarus border closure

https://www.theguardian.com/world/2025/sep/10/ukraine-war-briefing-russian-drone-alert-in-poland-...
2•Anon84•54m ago•0 comments

HarfBust: A font software puzzle for the curious

https://behdad.github.io/harfbust/
1•behdad•58m ago•1 comments

Odiff: The "fastest image comparison library in the world" was rewritten in Zig

https://github.com/dmtrKovalenko/odiff
2•signa11•1h ago•0 comments

Oakland AIs? The Ballers will be first professional sports team managed by AI

https://www.nytimes.com/athletic/6599707/2025/09/05/artificial-intelligence-baseball-manager-oakl...
1•t23•1h ago•1 comments

Chinese quantum processor rivals Google's Willow chip

https://www.livescience.com/technology/computing/china-achieves-quantum-supremacy-claim-with-new-...
3•rezaprima•1h ago•0 comments
Open in hackernews

Microsoft to Use AI from Anthropic in Partial Shift from OpenAI

https://www.reuters.com/business/microsoft-use-some-ai-anthropic-shift-openai-information-reports-2025-09-09/
7•andsoitis•4h ago

Comments

consumer451•4h ago
This is a really interesting development.

> Developers making Office AI features found Anthropic's latest models performed better than OpenAI in automating tasks such as financial functions in Excel or generating Powerpoint presentations based on instructions, the report said, citing one of the two people involved in the effort.

> Microsoft will pay its cloud rival Amazon Web Services to access the Anthropic models, according to the report. AWS is one of Anthropic's largest shareholders.

Any folks here who integrate LLMs into their products have opinions on when to use Anthropic's models over OpenAI's?

A_D_E_P_T•3h ago
In my experience, Anthropic's models may be slightly better for coding. (I think that GPT-5 Thinking/Pro are ahead, but I know that others disagree.) Where just about everything else is concerned, though, my impression is that they're way behind. They're particularly terrible at analyzing images. I recently gave GPT-5, o3, and Gemini 2.5 Pro an image analysis task (literally as simple as "describe this image"), and all of them passed the test with flying colors. I gave the exact same task to Opus 4.1 and it was spectacularly botched, twice.

So I might use Opus for coding and perhaps for agentic stuff, but I'd hesitate to use it for anything else.

al_borland•3h ago
The way things have been going, it seems like the only right move is to make the products AI model agnostic, where any model could plugin. There is too much flux in this space, and even based on user preferences, to be tied too tightly to any one model or vendor.
consumer451•3h ago
I used Vercel's ai sdk like everyone else, to try to stay agnostic. However, when I am actually working on a feature that uses an LLM API call, I find that each prompt might need tweaking specific to that model to get great results. That means if I switched models for that feature, I would likely have to that redo the prompting. Or, am I doing this all wrong?