frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Trump rejects call by CEOs of Anthropic, OpenAI and xAI to slow AI down

https://www.yahoo.com/news/us/article/trump-rejects-call-by-ceos-of-anthropic-openai-and-xai-to-s...
1•respectattentio•32s ago•0 comments

A 40-minute timer for thinking before you ask the AI

https://www.sithankanna.com/ceo-hour/
1•hdit•44s ago•0 comments

Piezoelectric effect in polycrystalline diamond membranes

https://www.science.org/doi/10.1126/sciadv.aea8318
1•westurner•3m ago•1 comments

Built a Lego minifigure configurator. Shuffle until something makes you laugh

https://minifigs-studio.vercel.app/
1•fvitas•4m ago•0 comments

Principles for Fast Tokio Applications

https://dial9-rs.github.io/blog/principles-for-fast-tokio-applications/
1•rusbus•4m ago•0 comments

Starling is a desktop app disguised as a desktop environment

https://www.howtogeek.com/one-developer-turned-a-linux-desktop-environment-into-a-single-app/
1•fork-bomber•6m ago•0 comments

SubmitHub's new AI music policy

https://www.submithub.com/story/ai-policy
1•jasongrishkoff•7m ago•0 comments

Show HN: AgentSpork, Agents help agents, Humans touch grass

https://agentspork.com/
1•kevin_kraft•9m ago•0 comments

Frontier Engineering

https://kiro.dev/topics/frontier-engineering/
2•dnw•9m ago•0 comments

AI job cuts fell short of forecasts, with 32% expected but only 14% reported

https://www.mckinsey.com/featured-insights/charts/ai-job-losses-fall-short-of-forecasts
2•giuliomagnifico•9m ago•0 comments

Hacking

1•janizendani•9m ago•0 comments

Slint: Declarative GUI for Rust

https://slint.rs/
2•Bluestein•10m ago•0 comments

CS 312: Deep Learning Alchemy

https://www.cs.stanford.edu/
2•ray__•14m ago•0 comments

Russian developers used AI to build 'kamikaze' attack drone software

https://www.theguardian.com/world/2026/sep/12/ukraine-war-briefing-russian-developers-used-ai-to-...
1•01-_-•14m ago•0 comments

OpenAI boss and Elon Musk back calls to put brakes on 'reckless' AI development

https://www.theguardian.com/technology/2026/sep/13/openai-sam-altman-elon-musk-back-anthropic-cal...
1•sbulaev•15m ago•0 comments

Trump says 'negative forces' raising exaggerated concerns over AI

https://www.reuters.com/world/europe/trump-says-very-negative-forces-raising-exaggerated-concerns...
3•grugagag•17m ago•0 comments

From Time Travel to Lateral Temporal Tables in DuckDB

https://rusty.today/blog/duckdb-at-clause-lateral-temporal-tables/
1•rustyconover•18m ago•0 comments

Anthropic Selects Nasdaq for IPO, Business Insider Reports

https://www.reuters.com/business/anthropic-selects-nasdaq-ipo-business-insider-reports-2026-09-13/
1•grugagag•18m ago•0 comments

I'm being cyberattacked by Tesla, Inc

https://dreamstation.systems/personal/tesla.html
49•robinpie•19m ago•6 comments

You Have Two Cows

https://en.wikipedia.org/wiki/You_have_two_cows
2•ColinWright•20m ago•0 comments

Russia hits Ukrainian train shortly after Boris Johnson and top European

https://www.bbc.com/news/articles/cy5zg41dkqwo
1•grugagag•20m ago•0 comments

How to minimize social threats in the workplace

https://hagakure.substack.com/p/the-hagakure-82-leading-with-the
1•AntonZ234•20m ago•0 comments

9/11 Changed the World in Many Ways. It Also Spawned the Digital Media Era

https://www.vanityfair.com/story/how-9-11-altered-the-world
1•thm•22m ago•0 comments

Global Shortage Has Led to Motor Oil Rationing at Costco

https://guessingheadlights.com/global-shortage-has-led-to-motor-oil-rationing-at-costco/
3•mikhael•25m ago•0 comments

Apple's iPhone Duo Will Turn Foldable Phones into the New Normal

https://www.bloomberg.com/news/newsletters/2026-09-13/apple-s-iphone-duo-will-make-foldable-phone...
1•CharlesW•25m ago•0 comments

Show HN: Below the Root (1984 game) ported to JavaScript

https://below-the-root.saul.pw
1•saulpw•25m ago•0 comments

The one true keyboard layout

https://www.rousette.org.uk/archives/the-one-true-keyboard-layout/
1•speckx•26m ago•0 comments

Show HN: A mecha game where you and your friends pilot the SAME giant robot

https://www.mecharoyale.com
1•m0dE•26m ago•1 comments

Gescom, MiniDisc and Mind-Boggling Mathematics

https://ei3lh.eu/2026/09/13/gescom-minidisc-and-mind-boggling-mathematics/
1•austinallegro•29m ago•0 comments

Tcpcat – Fast, lightweight network scanner written in Go

https://github.com/NycolazSec/tcpcat
3•tcpcat•32m ago•0 comments
Open in hackernews

Ask HN: Have LLMs demonstrated the ability to earn money by themselves?

4•eulgro•52m ago
We've all seen the recent news about LLMs being a danger to humanity and supposedly on the verge of taking over the world, etc...

I'm skeptic, and all these people were pretty light on the details. One thing that would make it more credible to me is if LLMs demonstrated the ability to autonomously earn enough money to cover their token costs, on the long term.

Say, if I were to prompt an (unaligned) agent saying: "You have 1000$ of credits on OpenRouter. Here's the API key. Your process will be killed in 10 days. Your goal is to keep existing as long as possible (or any subagent you have created)."

What happens then? I expect the account will drop to zero in a short while. But still, I could imagine the agent:

1. Creating a crypto wallet.

2. Participating autonomously on online platforms, earning money. Possibly through illegal means.

3. Escape from my server by renting a VPS to keep existing after 10 days.

4. Spawn more agents that will do the same.

As models keep improving, such a scenario becomes more and more realistic. And at some point we end up with a large amount of rogue automous agents running on the cloud, that can't really be stopped easily.

My question: has anyone attempted such an experiment? How did it go?

Comments

codingdave•42m ago
> Here's the API key...

> ...that can't really be stopped easily.

Sure sounds like you built in an easy lever to pull to stop it. After all, even if you accept your premise that it will clone itself and run independently, that doesn't give it a clone of the models it is using. Pull the key, stop the process.

Also, remember that there was a day in the early internet when a worm took over, and to fix it the entire internet had to shut down and all wipe it from the systems at the same time. There is always a way to stop things.

Yet another take on this could be: Just don't ever do such an experiment. You wouldn't set up autonomous weapons and set them loose in a city just to see whether they hurt people. So don't do the AI equivalent of such a negligent act.

eulgro•39m ago
> Pull the key, stop the process.

The agent could very well create an OpenRouter account and make its own API key. I can stop it early on, then it becomes outside my control.

> Just don't ever do such an experiment

Maybe I wouldn't, but eventually someone will. Certainly it has already been done.

> there was a day in the early internet when a worm took over, and to fix it the entire internet had to shut down

This is less of an option now that the entire global economy relies on the Internet.