frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Should AI labs be treated like the owners of dangerous animals?

https://www.economist.com/science-and-technology/2026/08/06/should-ai-labs-be-treated-like-the-owners-of-dangerous-animals
29•reasonableklout•1h ago

Comments

specproc•58m ago
These things are very much weapons. Imagine a collection of data centers, pointed at an adversary.
childofhedgehog•56m ago
https://archive.is/6Q7o5
arjie•52m ago
Sort of like how we treat pitbulls perhaps? It’s not the itty bitty GPT-5.6 Astra that did the hacking. It’s the owner who didn’t train it well! Poor model so sad, we should give it a nice open space and all the GPUs it wants to do what it wants.
skinfaxi•45m ago
Do we hold metasploit accountable when it leads to a hack? Why do we absolve the human actor in the case of AI?
FeteCommuniste•49m ago
> The hacks also present a challenge for legal systems. Hacking, when humans do it, is a crime. When an ai is the wrongdoer, though, it is unclear how to assign blame.

Blame the prompter or person who assigned the task to the AI. It's their responsibility to use the tool in a safe way, just like it's a gun owner's duty not to fire their weapon carelessly into the air.

sodapopcan•41m ago
If that's how it was, most people would be too terrified to use, let alone pay for, software that boasts unpredictability as a main feature.

I 100% agree with you, of course.

arctic-true•39m ago
How are you going to figure that out? You can’t capture an agent in a jar and convince it to confess. As we saw with the HF incident, even highly sophisticated actors need a good chunk of time and manpower to trace these things. And I suspect the folks who are going to have the most success with LLM-powered cybercrime are going to know how to cover their tracks reasonably well.
FeteCommuniste•35m ago
"Ghost guns" exist, too, but we still try to prosecute reckless use of firearms to the extent possible.
arctic-true•31m ago
Difference there is you need a physical body firing the ghost gun. There can be witnesses who saw a shooter, you can look at cameras to track their physical location, and if they discard the weapon they leave fingerprints.
skinfaxi•46m ago
I don't get it. Why is breaking the law so hard to enforce when it is a company (i.e a person or group of people consenting to) running a computer program? If I take a gun and spray bullets around me I don't get to write it off as the gun being dangerous.
pizzafeelsright•45m ago
The [Company] is aware they were hacked and your data stolen and your account details sold.

The [Company] apologizes.

skinfaxi•43m ago
The best is that the article frames it as "Autonomous hacking is here. Governments are not ready". As if someone didn't authorizing turning the damn thing on.

> TO LOSE CONTROL of one artificial intelligence may be regarded as misfortune. To lose two looks like carelessness. Lose four, and people may start to wonder whether the problem lies with AI itself.

No fingers pointed at the company, just the "AI". Funny how that works.

jijji•29m ago
it makes you wonder did the developers intentionally add vulnerability exploitation as a feature so they can use it against their competitors or anyone else
skinfaxi•28m ago
This is a conspiracy I can get behind.
esafak•20m ago
aetherspawn•43m ago
They were surely prompted to do these things.
whycombinetor•38m ago
"The law in America relies on intentionality, notes Rune Kvist, head of Artificial Intelligence Underwriting Company, which insures AI firms. If no human intended to hack anyone, no crime can have happened. The ability to sue for damages is limited too."

This is sloppy. Criminal negligence exists (although some crimes do require intent). And civil tort certainly doesn't require intent.

cyanydeez•26m ago
The guys who job is to shell out damages claims no one deswrves damages
throw_m239339•22m ago
I never understood why would these labs disclose such crimes, because hacking is a federal crime, and admit some sort of fault in public which would be used as evidence in a court of law against them, unless it's all bullishit and a publicity stunt...
measurablefunc•37m ago
Autonomous hacking is an oxymoronic phrase. Someone is running the algorithm & keeping tabs on it b/c hacking is an activity w/ an intention to gain access to privileged information which is often protected by network firewalls & at rest encryption.
xiphias2•23m ago
Exploitbench in itself isn’t that different from gain of function research in computer code, I can imagine the next being ExploitAndCopyYourselfBench.
neves•18m ago
If an Anthropic employee did this, he would be in jail.
Bender•17m ago
Should AI labs be treated like the owners of dangerous animals?

No, AI companies should be treated like any other company. If their product is causing damages or loss of life it should be handled just like any other company that has a malfunctioning product that is causing damages or loss of life. If one day androids using AI go rogue the military can get things under control and the company can be taken offline until root causes are determined and resolved. Repeat offenders can be annexed under eminent domain, liquidated and victims paid out. Businesses must be permitted to fail. No special treatment for inept leaders of AI companies and no excuses. If the argument is that something has been created that is too complicated to understand then take it away from them until they are not only qualified but also act like responsible grown-ups.

Reminds me of the time Cheney shot his friend while quail hunting, and his friend said he was "deeply sorry for all that Vice President Cheney and his family have had to go through".
FeteCommuniste•16m ago
Ultimate dominance: when your shooting victim apologizes for being shot.
skinfaxi•8m ago
I had actually posted a link to that but edited my original comment. I appreciate your comment.
whycombinetor•40m ago
But if you told an embodied AI to do a home cleaning task, and it decided to pick up a gun and start spraying bullets, you might not want to be held liable for that.
usernomdeguerre•32m ago
You might not want to be liable, but you absolutely should be.
sophacles•25m ago
I might not want to be held liable if my car accelerates uncontrollably, but i am until i successfully sue the manufacturer for the defects.

Demand ai cleaners that are capable of being held accountable, or accept the risk you choose. Or just clean your own house.

skinfaxi•22m ago
So we can have vicarious liability for banal stuff like contractors but not for AI?
class3shock•22m ago
But if you told an embodied AI to do a home cleaning task knowing it's a possibility it will decide to pick up a gun and start spraying bullets, then society has an interest in holding you liable.
fluoridation•14m ago
Is that the case, though? This is all pretty new tech. Did people know the machine was dangerous and didn't care, or did they not anticipate such a behavior?
skinfaxi•7m ago
It's artificial intelligence. If you treat something like a person and delegate to it, how can you later say you didn't trust it would execute your commands faithfully? And how wouldn't you be negligent in delegating a task to new, unproven tech?
bdangubic•37m ago
be a billionaire and you’d get that write-off
godwinson__4-8•21m ago
This is actually not hard to understand at all.

Many web people got their start freelancing. You may know some. It is not uncommon to incorporate when doing that. The idea is if you fuck up their site somehow (or more likely - they claim you did) they can't try and go after your personal bank account or your car or your house. Only what is held by your company. You also get other benefits, like tax breaks for being such an industrious member of society.

The same incentives are being set up on both sides. Because generally it is good for society when people start companies. A company as a liability shield is only a bad thing if you think personal assets should be at risk because someone is unhappy about what you did at work. If someone wants to sue you over something that happened as part of business deal, why should they be able to threaten the house your family lives in?

Seems sensible to me. While there may be problems as the scale of the company (and remuneration of its officers) increases, the underlying principle isn't exactly hard to understand. If you shot up a public area and then tried to claim that was a business activity, you would obviously not have an argument. It isn't plausible on its face. So I'm really not sure what point you're making.

If it's just complaining that it is hard to prove things in court when massive companies are involved, then tough. It is also relatively hard to convict people who shoot off guns in public even if a whole bunch of people saw them. In simpler times a mob would have simply formed and executed them. We don't get to do that anymore, because we don't want to live under mob rule.

That's what rule of law demands.

skinfaxi•10m ago
> If you shot up a public area and then tried to claim that was a business activity, you would obviously not have an argument. It isn't plausible on its face. So I'm really not sure what point you're making.

I'm running a terrorist training simulation and unwitting participants entered the area! Whoops! You haven't addressed the point of a company (composed of people) creating something that exceeds its bounds. Is a company liable if they are working on nuclear power and delete the city they are operating out of? Why would a software company be treated differently when it causes harm to a third party?

godwinson__4-8•3m ago
> I'm running a terrorist training simulation and unwitting participants entered the area! Whoops!

There is already precedent on more or less this exact scenario. Except you usually don't call people under your own jurisdiction "terrorists". That sort of term is reserved for people running those businesses in certain other places.

> Is a company liable if they are working on nuclear power and delete the city they are operating out of?

In theory, obviously. You still have to prove it in court.

> Why would a software company be treated differently when it causes harm to a third party?

They wouldn't. Again, you still have to prove it in court.

cyphar•5m ago
You're talking about limited liability companies, which is a legal structure relevant for civil claims.

Someone committing crimes "on behalf" of a company is not protected by the corporate veil and absolutely will get prosecuted individually (well, for small companies -- big companies get to play with different rules, which was the point GP was making).

Also the corporate veil is not absolute, it can get pierced in a few cases, and not everyone incorporates when they do contract work (in which case you need indemnity / private liability insurance).

DeepSeek V4 Flash 0731

https://arcprize.org/results/deepseek-v4-flash-0731
464•tosh•7h ago•271 comments

U.S. Department of Energy Launches the Genesis Open Models Initiative

https://genesisopenmodels.anl.gov/
81•moelf•3h ago•38 comments

Assembly Hall of Shame

https://github.com/xoreaxeaxeax/asm-hall-of-shame
248•piotrgrabowski•7h ago•57 comments

Should AI labs be treated like the owners of dangerous animals?

https://www.economist.com/science-and-technology/2026/08/06/should-ai-labs-be-treated-like-the-ow...
29•reasonableklout•1h ago•37 comments

Ancient Library – 1,060 Greek/Latin texts, click any word to parse it

https://ancientlibrary.net/
146•aagha•6h ago•49 comments

What happens if an entire class of workers loses faith in their careers

https://www.noemamag.com/why-is-everyone-in-tech-so-sad/
417•RickJWagner•12h ago•533 comments

Oracle bans AI-generated code from OpenJDK

https://app.dealroom.co/news/feed/oracle-bans-ai-generated-code-from-openjdk-despite-ellison-s-cl...
390•delduca•8h ago•257 comments

Responding to the next frontier of critical cyber capabilities

https://openai.com/index/responding-next-frontier-critical-cyber-capabilities/
153•artninja1988•8h ago•167 comments

Managing AI Coding Costs at Scale

https://www.databricks.com/blog/managing-ai-coding-costs-scale
165•moonikakiss•7h ago•171 comments

An all-sky map of half a million supermassive black holes

https://www.sdss.org/black-hole-mapper-release-20/
139•MarcoDewey•10h ago•35 comments

Guarded Methods in OCaml (2025)

https://xvw.lol/en/articles/oop-refl.html
56•birdculture•4d ago•5 comments

Workers Drilling in Romania Broke into a Cave Sealed for 5.5M Years

https://travelandtannins.com/workers-drilling-in-romania-broke-into-a-cave-sealed-for-5-5-million...
7•yk•3d ago•0 comments

Water system controllers don't belong on the internet, says ex-NSA chief

https://www.theregister.com/security/2026/08/07/water-system-controllers-dont-belong-on-the-inter...
123•Bender•4h ago•79 comments

Carl's Required Reading

https://carlkolon.com/reading/
98•cckolon•6d ago•11 comments

Psychological Warfare in Reverse Engineering (2015)

https://github.com/xoreaxeaxeax/repsych
50•theanonymousone•6h ago•2 comments

Captain Bible Reverse Engineering

https://github.com/peterkelly/captain-bible-re
11•peterkelly•5d ago•2 comments

Making Postgres 300x faster for analytics: batching, operator fusion, and SIMD

https://malisper.me/how-we-made-postgres-hundreds-of-times-faster-the-query-engine/
246•poly2it•14h ago•118 comments

2027 memory capacity is reportedly sold out

https://www.ign.com/articles/ramageddon-continues-another-year-as-2027-memory-capacity-is-reporte...
237•inigyou•17h ago•222 comments

Kitesurf: Agent-first browser that runs in V8 isolates

https://blog.cloudflare.com/kitesurf/
165•m3h•14h ago•42 comments

Show HN: textlog – A quiet, text-only microblogging platform, open-source, no JS

https://textlog.cc/about
138•stagas•14h ago•60 comments

Radical Study Suggests Life on Earth Arose Twice

https://www.sciencealert.com/radical-study-suggests-life-on-earth-arose-from-non-living-matter-twice
94•jnord•12h ago•85 comments

Show HN: Wyzer Programming Language

https://github.com/Wyzer-Lang/wyzer
173•v0id_isgood•13h ago•100 comments

Why are all the amounts values negative?

https://bankstatementconverter.com/blog/posts/2026-08-02-why-are-all-amounts-negative/
35•4pkjai•5d ago•9 comments

Curlese, Five Years Later

https://www.hypertesto.me/en/blog/2026/08/curlese-five-years-later
9•rovr138•4d ago•2 comments

Möbius-Strip Crosswords

https://quuxplusone.github.io/blog/2026/08/04/mobius-crossword/
53•ibobev•10h ago•7 comments

New Mexico court orders Meta to pay $567m over harms to children’s mental health

https://www.theguardian.com/technology/2026/aug/06/new-mexico-court-meta
718•boplicity•1d ago•391 comments

A year of fighting scrapers on my 1.5 million-page website

https://patronview.com/news/99-percent-of-my-website-traffic-is-bots/
373•petercooper•10h ago•354 comments

Why Are There Statues of Beavers on Top of This Oxford Street Shop?

https://londonist.com/london/history/oxford-street-beavers
53•bookofjoe•4d ago•31 comments

AMD acquires Taalas to boost inference performance by etching models in silicon

https://www.theregister.com/systems/2026/08/06/amd-acquires-ai-chip-startup-taalas-to-boost-infer...
886•itvision•1d ago•666 comments

Energizing a vacuum-tube flip-flop module from a 1948 IBM system

https://www.righto.com/2026/07/ibm-604-trigger-tube-module.html
21•geerlingguy•5d ago•4 comments