frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Captive Scam Workers in Cambodia Couldn't Flee as Bombs Fell

https://www.wsj.com/world/asia/captive-scam-workers-in-cambodia-couldnt-flee-as-bombs-fell-ed067875
1•perihelions•2m ago•0 comments

GNU Taler v1.3 Released

https://www.taler.net/en/news/2025-13.html
1•F3nd0•3m ago•0 comments

OSI Did Initiate Open Source

https://openpath.quest/2025/osi-really-did-initiate-open-source/
1•whit537•5m ago•0 comments

OxCon 2025: Update on Update (great talk) [video]

https://www.youtube.com/watch?v=M-ZLz8Wg34s
1•bags43•7m ago•0 comments

How Willie Nelson Sees America

https://www.newyorker.com/magazine/2025/12/29/willie-nelson-profile
1•NaOH•9m ago•0 comments

Germany's far-right AfD accused of gathering information for the Kremlin

https://www.politico.eu/article/germanys-far-right-afd-accused-of-gathering-information-for-the-k...
2•robtherobber•11m ago•0 comments

The Military's Social Media Purge

https://www.lawfaremedia.org/article/the-military-s-social-media-purge
1•hn_acker•11m ago•0 comments

FOSDEM 2026 – Schedule

https://fosdem.org/2026/schedule/
2•Tomte•12m ago•0 comments

The Recursive Lemniscate: A New Topology of Human-AI Alignment

https://renormalize.substack.com/p/humanistic-ai-a-vision-for-the-21st
1•getnormality•12m ago•1 comments

CBS 60 minutes on CECOT only aired in Canada

https://archive.org/details/cecot-60mins
7•clacker-o-matic•13m ago•0 comments

How the Catholic Church thinks about superintelligence

https://www.transformernews.ai/p/paolo-benanti-catholic-church-vatican-superintelligence-artifici...
2•atlasunshrugged•14m ago•0 comments

Show HN: Textbooks are hard. I made AxisY to visualize complex topics

https://axisy.tech
1•superhuang•14m ago•1 comments

Ask HN: Would you use % based affiliate software instead of monthly fee?

1•StanMakesThings•15m ago•0 comments

Show HN: I Built a US Grid Telemetry API Using Azure Functions and Python

https://github.com/Norris-Eng/gridwatch-kill-switch
1•Norris-Eng•15m ago•1 comments

Ask HN: What data are you sharing with LLMs?

1•elevatortrim•16m ago•0 comments

An Interesting Year

https://werd.io/2025-in-review-an-interesting-year/
1•benwerd•16m ago•0 comments

Reverse Engineering a Mysterious UDP Stream in My Hotel (2016)

https://www.gkbrk.com/hotel-music
4•bayesnet•18m ago•0 comments

Typing Considered Harmful: Why Voice Coding Works

https://utter.to/blog/voice-based-coding-and-meta-prompting
1•helro•18m ago•1 comments

Memelang: An Axial Grammar for LLM-Generated Vector-Relational Queries

https://arxiv.org/abs/2512.17967
1•bri-holt•18m ago•0 comments

Show HN: Tokscale – See who's burning the most tokens across all platforms

https://github.com/junhoyeo/tokscale
2•junhoyeo•18m ago•0 comments

We Killed RAG, MCP, and Agentic Loops. Here's What Happened

https://www.decodingai.com/p/building-vertical-ai-agents-case-study-1
1•pauliusztin•19m ago•0 comments

Corsair sent me DDR5 RAM Replacement Sticks and the new RAM looks like this?

https://old.reddit.com/r/pcmasterrace/comments/1pt20hk/corsair_sent_me_ddr5_ram_replacement_stick...
3•croes•23m ago•0 comments

Piston-Driven Syringe-Extrusion Bioprinter Using an FDM Frame

https://www.mdpi.com/2313-7673/10/12/811
2•PaulHoule•23m ago•0 comments

Show HN: A CLI for ADHD Productivity, Aggregates Gmail, Calendar, GitHub

https://github.com/mduffster/utility-explorer
3•dwmd14•24m ago•0 comments

Experts Warn U.S. Power Grid Is at Risk of Catastrophic Failure

https://madgewaggy.blogspot.com/2025/12/experts-warning-about-us-power-grid-is.html
1•SanjayMehta•24m ago•1 comments

Apex: A unified Markdown processor for all major flavors in one C tool

https://github.com/ApexMarkdown/apex
1•thecosas•26m ago•0 comments

Xerox and Stack Overflow partnered to preserve knowledge and power innovation

https://stackoverflow.blog/2025/12/23/how-xerox-and-stack-overflow-partnered-to-preserve-knowledg...
1•mikece•27m ago•0 comments

Netflix buys Estonian avatar pioneer Ready Player Me

https://estonianworld.com/business/netflix-buys-estonian-avatar-pioneer-ready-player-me/
1•atlasunshrugged•27m ago•0 comments

39C3 Schedule (Fahrplan)

https://fahrplan.events.ccc.de/congress/2025/fahrplan/
2•qwertox•27m ago•0 comments

Building this platform for CTO's/devs/founders

1•akhnid•28m ago•1 comments
Open in hackernews

Ask HN: Can You Patent Prompts?

3•rcarmo•2h ago
I decided to ask this question because of a rather contrived informal discussion I'm involved in with colleagues from both academia and industry, neither of which are lawyers but who share a few concerns about how we can go about creating new "things" and intellectual property in an increasingly AI-centric world.

You can take prompts as "processes", or "embodiments" or any variation on intellectual property, but I'd love to know what people think about this--my particular concern was that if you ship your software as Open Source, could someone nitpick over your prompts and claim they were taken from their IP/software? How valid would this be given that prompts are, essentially language? How detailed does a prompt need to be to be complex intellectual property that describes a process? Can we consider it as code and an embodiment of a concept?

Discuss. And Happy Holidays!

Comments

KellyCriterion•2h ago
Excellent question - but I'm not a lawyer!

I want to add my question: How can something be a patent, if the result of the prompts are not always the same on each run? My understanding is: Only things can go into patents, which are "reproduceable"?

Maybe Im wrong?

Mountain_Skies•1h ago
Along those lines, if the model's output was actually deterministic, would the patent be bound to using the prompt with only one particular model since the output would be different, in some cases wildly different, with other models or even other versions of the same model?

My absolute wild guess as a layman is that prompts will at most get "trade secret" protection and won't be able to be copyrighted or patented.

theflyestpilot•2h ago
Hmmm, I think of prompts more like recipes than an algorithm. Similar in both regards, but to the layman, I think prompts are becoming as common practice in daily lives as recipes.

Recipes can be trade secrets but not IP.(Unless it's a super rare circumstance that proves uniqueness in a 'Food Science' way).

I surely hope no one can patent a prompt. That would be an annoying world to live in.

KellyCriterion•1h ago
Recipes can be somehow a brand: the name "Original Sacher Torte" is protected and can only be used by Hotel Sacher, but the recipe itself isn't
stop50•1h ago
its just an brand like you can register yourself one.