frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Accept 'bad things' in return for benefits of AI, says Sam Altman

https://www.theguardian.com/technology/2026/oct/05/sam-altman-open-ai-chatgpt-benefits-risks
30•samizdis•46m ago

Comments

2OEH8eoCRo0•44m ago
Bad things like what? Armed psychopaths targeting AI executives?
jeanpah•38m ago
Good things like what? Mass unemployment and concentration of power?
Joker_vD•35m ago
It really is the best of all the possible worlds, you know.
DarkWiiPlayer•27m ago
Why just the psychopaths? What about the risk of perfectly healthy, reasonable armed people targeting AI executives?
ejee•23m ago
I sometimes wonder if the optimal outcome is - the industry leaders have at it and go for accelerated progress but accept they then place themselves in danger.

I have a lot of interest in seeing how the common person response in this scenario.

GolfPopper•14m ago
With the increasing development of capable drones, I keep wondering how long it's going to be before we start seeing people who have given up on law and regulation resorting to drone-enacted vigilantism.
jackb4040•17m ago
Targeting? I think you meant to say "upstanding citizen in the vicinity when chemical reaction leads to AI-CEO-involved incident"
hagbard_c•11m ago
Don't go there. One of the things separating functional states - warts and all - from failed ones is that the populace does not turn to deadly violence to settle disagreements and griefs. Yes, there are people who shoot/stab/${kill_method} other people but they are prosecuted as criminals, i.e. the opposite of those perfectly healthy, reasonable armed people you mention.

Increased acceptance of political violence begets more political violence which in turn seems to increase its acceptance, see e.g. the killing of the healthcare CEO and subsequent arrest of the killer which significantly increased "democrats" support for political violence [1] so be careful what you ask for.

[1 https://academic.oup.com/poq/advance-article/doi/10.1093/poq...

mdp2021•6m ago
> separating functional states - warts and all - from failed ones

Systemic failure triggers the parasystemic shortcuts. Lack of justice and fairness has grave consequences - they must be curbed with justice and fairness (and their roots) exactly.

hagbard_c•4m ago
OK, cowardly down-voter, please explain why you think I'm wrong here instead of just hiding behind that button. This is a discussion site, not a Party Congress, we're here to discuss things. As it stands now I can only assume you want more political violence. Do you? If so, why?
harimau777•3m ago
What's the alternative? The system is completely nonresponsive to the poor and oppressed. Right or wrong, it is only natural that people turn to violence when they have no other option.
GolfPopper•19m ago
The AI executives are the psychopaths.
LoganDark•39m ago
No, we should not "accept" bad things. We can look forward to benefits, but BENEFITS ARE NOT AN EXCUSE FOR THE BAD THINGS. Ever.

Edit: I mean that we should continue to shun antisocial behavior even when it involves LLMs, not that we should blame the technology. The technology isn't an excuse, no technology should ever be an excuse -- that's what I mean by not "accepting" bad things. I don't mean that LLMs are not "worth it" thanks to the abuse, just that we should continue to treat abuse as such rather than giving anybody a free pass.

RealityVoid•37m ago
This is kind of how the world works. We have a set of tradeoffs to make and the downside of the tradeoff we accept in exchange of the advantages.
herrherrmann•33m ago
Not really. Rich entitled tech bros don’t dictate how the world works. There is lots of progress possible without needing the vast majority of people to pay a huge price for it.
RealityVoid•23m ago
I've said nothing about who dictates. I have said that everything in life is a tradeoff and making a tradeoff requires talking the bad with the good. It's effectively something we can't escape. If we want to do this tradeoff with AI, I did not take a positions as well. But I posit there will be tradeoffs.
LoganDark•31m ago
No, the benefits of AI do not mean anything like we have to accept AI slop contributions to open-source projects. We can still shun the blatant abuse. We don't have to just shut up about it because we might stop frontier development. We won't.

To me, "accept" means to socially tolerate it. No. Bad. It's not socially tolerable. It is still wrong to use LLMs for evil. I don't think that means LLMs are a fundamentally evil technology, but that does NOT mean anyone should get a free pass to use them FOR evil.

mschuster91•38m ago
Yeah thanks but no thanks. Society does not have to accept "bad things". Society can pass laws to prohibit bad things.

And society will pass laws. Europe has already begun doing so, and the US will once this goddamn corrupt administration is gone. People are already protesting against AI, especially against datacenters and surveillance like Flock. It's only a matter of time until individual actors end up in the focus of protests.

omnicognate•32m ago
I was with you up to the last sentence.
mschuster91•20m ago
Think of society as a pressure cooker and Sam Altman, OpenAI etc. as the oven's fire.

There's now just three possible futures:

- the fire is extinguished, aka the AI actors go bust. Nothing bad happens. Well, a bunch of rich people will be less rich, but society as a whole will be somewhat all right.

- the relief valve (society passing laws) is activated, pressure escapes.

- there is no pressure relief. Eventually, the pressure cooker bursts, aka revolution.

tikimcfee•31m ago
You are correct. I appreciate your willingness to say it's societal and therefore by construction political. And, the idea that "we accept the bad things" is so prevalent in this community that it's baffling. I have been more vocal (and emotionally representing, to admit) about how this community is willing to condemn the abuse of privacy, but then share excitement about agents and LLM work from Facebook and the like. The reasoning is usual, "yeah well someone is going to do it," or, "well it's hard or impossible to make people not use it, so what's your solution?" I've been finding that mentality is both a blockade to progress, and maddening to boot. If those same minds instead did something about it by sharing the difficulty with others, proselytizing the importance of sharing your opinion about what the bad thing is, and then standing by it whether it's hard or not - that would actually help things be not so bad.
trollbridge•32m ago
Note that the regulation he’s pushing back against is simply having third party defined and reviewed guardrails, as opposed to “just trust us bro”.
jackb4040•21m ago
They shift between

"we need the government to halt model training for us (and potential competitors who are currently behind us) due to hypothetical sci-fi future bad things"

on the one hand, and then

"we should not face consequences for that time we helped a teenager plan a mass shooting"

on the other so effortlessly, so instantaneously. And virtually the entire media helps them do it!

Hamuko•31m ago
I'll consider it when Sam starts accepting getting Molotov cocktails thrown at his house. Surely we all need to accept bad things, right?
DarkWiiPlayer•30m ago
I believe the world should accept the "bad things" that result from allowing me to take other people's property, because of all the good I'd do with the money I'd take from the rich.

Hear me out, I know common sense says theft is bad, but just think about the hungry children I could feed if the world just agreed to let me take whatever I wanted.

...except AI won't even feed hungry children, it will only make the rich richer and flood the internet with slop.

reallymental•30m ago
Interesting, I wonder what the public's opinion would be if every plastic/chemical maker out there says the exact same thing.
mdp2021•22m ago
...They do, implicitly? If the "public opinion" changed perspective in either way after such statement, they would have missed the facts.
monideas•12m ago
There are thousands of active plastic and chemicals plants in the US and yes occasionally there are accidents, yet there are still thousands of active plastic and chemical plants in the US.
fourseventy•30m ago
Every technology advancement comes with good things and bad things. Fire - Good: can cook food, Bad: can be used for war.
agos•29m ago
If fire was invented today, it would be a subscription
DarkWiiPlayer•25m ago
> If fire was invented today

You know you're starting to be personally affected by the decline of civilization when your brain can conjure up the notion of fire being "invented".

mdp2021•18m ago
"You know there is trouble when reality surpasses fantasy" (~~ Berke Breathed)

We know we are in trouble.

fwlr•19m ago
OpenFlame’s latest model has really good scores on FahrenBench but I’m having a lot of trouble keeping it contained. I use a proper firepit but it keeps spreading to nearby homes and vehicles.
harimau777•21m ago
That doesn't mean that the good and bad things are equitably distributed, that the good and bad things happen at the same time, or that the bad outweighs the good. All three issues are at play here: The benefits are likely to disproportionately help the rich and the pain is likely to disproportionately help the poor. New prosperity is likely to come too late for the people whose lives are being disrupted. It seems likely that the benefits of culture being consumed by AI slop outweighs the benefits (I'd like to be more detailed, but honeslty it isn't completely clear what the benefits are).
neonnoodle•29m ago
'Bad things' [for you] in return for benefits [for me].
palmotea•27m ago
"Some of you may die, but that's a sacrifice I'm willing to make."
teekert•24m ago
"All animals are equal, but some animals are more equal than others"
datsci_est_2015•24m ago
I’ll add one to the pile:

“Socialize the costs, privatize the profits.”

delis-thumbs-7e•28m ago
I think the AI provider whose model does “bad things” should be held responsible of such things, going to a “bad place” - such as a prison - if the “bad things” are bad enough. Let’s see how that affects your IPO.

E: typos

jameskilton•27m ago
I will accept the "bad things" if and only when Sam Altman exposes himself to those bad things first e.g., living next door to an AI data center, being subjected to 24/7 surveillance, and having everything he says and does scrutinized for illegal activity, ticketing and jailing him appropriately.

Until that happens, no. Sam Altman is here telling us to accept the cruelty of his actions and his company so that he can be more wealthy.

delis-thumbs-7e•15m ago
I don’t think he needs or wants any more money. These guys already have so much moolah that it is virtually meaningless to them. They want power. Remember when Zuck tried to run for president?
mdp2021•14m ago
> being subjected to 24/7 surveillance, and having everything he says and does scrutinized for illegal activity, ticketing and jailing him appropriately

That thing has been rising for decades - why should Sam Altman now become the culprit and scapegoat?

That of surveillance and the use of pattern matching is a legislative matter - not a matter of regulating RnD.

Lerc•27m ago
Is this essentially a claim of the form that we should accept that car accidents happen if we are to drive cars?

It is certainly pitched as if it is a claim we should be accepting of individual instances. Which is not the same as accepting that they occur.

There's a big difference between acknowledging that something cannot be universally beneficial and claiming that the bad aspects are inconsequential.

tarkin2•22m ago
Is that title absent of meaning? Or at least highly misleading.

Every invention has positives and negatives. But no inventor or factory is at liberty to force the negatives upon society without regulation, or in Sam's case without being subject to the laws on theft and hacking.

He may as well come out and say "the ends justify the means" and the ends of course benefit Sam, and those subject to injustice, or negative externalities, should be happy that they can pay Sam for a subscription.

Or perhaps "Look, yes I stole, and yes I hacked, but now I've got this amazing tool you can use for free, and when you use it, and give me money for the full version, it will eventually replace you in the workplace and it will reduce you and your kids intellect after you outsource your thinking to it, and become dependent on it, which will be fine because you can just use the tool even more to make up for that."

Of course with a president who believes him to be above the law and American institutions, is it any wonder the people at the top of American society also believe it to be true? Not to mention their cheerleaders in the populous.

mdp2021•10m ago
> Is that a sentence absent of meaning? Or at least highly misleading.

Do you mean:

> “I wouldn’t take a trade of saying we will make sure there’s no major hacks, there’s no misuse of this technology, there’s zero scams or all the other bad things that will happen because I think that people will do tremendously – orders of magnitude – more good stuff than bad stuff.”

metalman•19m ago
someone could, as a favor, tell them to check the best before date on the rationalists koolaid, or maybe go back to the old formula that had a lot more woooo in it, in any case it's still too early on a monday to be indulging so heavily.
snazz•18m ago
> “There’s about a 50% chance we all die because of the development of smarter-than-human AI systems,” he said, adding that actions taken over the next two to 10 years would be decisive.

What’s the mechanism by which this happens? I feel like people making these predictions live on a different planet.

harimau777•11m ago
It seems notable to me that the tech bros ignore the obvious solution:

If we assume that they are right: AI is inevitable and certain significant negatives cannot be avoided, then the obvious solution is for society to put together a social safety net for those who are effected. Altman could easily decide to campaign for and donate to a party which made serious strides towards establishing such a safety net.

The fact that they ignore the obvious solution suggests to me that they aren't quite being honest about their motives.

pupppet•2m ago
And if the bad things happen, just mosey on down to your private island or doomsday bunker.
avaer•1m ago
Most on HN are in a privileged position towards the "haves" vs "have nots" of AI, so maybe there is "benefit" for us if it further stratifies us into the "haves".

But for the average person, I can't really think of a single net benefit except porn (even then arguable) and porn-adjacent things like roleplay. And AI companies don't even want you to have that.

Unless you are already in a privileged class, AI isn't reducing your working hours, it's not putting more money in your pocket, it's not adding anything to your life. The average person is not going to accept "bad things" when they are truly getting nothing in return.

nailer•30m ago
https://yourlogicalfallacyis.com/strawman
LoganDark•29m ago
They're twisting my definition of the word "accept". I was clarifying what I meant with an example. I think instances of evil behavior need to be shunned regardless of the other benefits of the technologies being used for evil. No one should get a free pass to do evil based on the technology.

I also believe the technology itself deserves to be improved so that it can be used for good. That doesn't mean I want more evil or to be more accepting towards evil. Evil should still be shunned.

harimau777•26m ago
Can you expand on this? It is not clear what part of the post you regard as attacking a strawman.
LoganDark•24m ago
It probably looked like I was equating "accepting the tradeoffs" to "condoning LLM slop contributions to open source projects". I was referring to enabling individual instances of abuse though, not accepting the technology overall in spite of the bad behavior. It's enabling the individual instances that I'm protesting, because I firmly believe we shouldn't. I should've been clearer, my bad.
iAMkenough•31m ago
Exactly. We accept the downside of being monitored and manipulated by the billionaire class in order to enjoy the fruits of social media.
ejee•24m ago
Trade offs start with comparing two goods btw.

I suppose monitoring is a good in the eyes of those who influence society.

iAMkenough•23m ago
Social media isn’t a good? I don’t understand your point.
harimau777•29m ago
Who gets to make that decision?
RealityVoid•25m ago
Nobody and everyone. Usually the incentives and local constraints and culture meld to have people in an area converge on a set of practices.
harimau777•19m ago
I don't think that's how it actually works in practice. In practice it is dramatically more likely that the rich and powerful get to decide who makes the tradeoffs. That is dangerous in normal times but especially considering the revealed nature and priorities of the tech bros and fascists currently in power.
palmotea•24m ago
> Who gets to make that decision?

Sam Altman, Dario Amodei, Elon Musk, David Sacks, etc. All the rest of you proles can fuck off. Stay in your lane.

ejee•26m ago
Nice try. Try going deeper.

Mollie strengthens Pay by Bank to power the next generation of European payments

https://www.mollie.com/news/mollie-relaunches-pay-by-bank
1•HelloUsername•1m ago•0 comments

Our new deal We stopped being an agency, and became a platform

https://feed.garden3d.net/p/our-new-deal
1•rudenoise•1m ago•0 comments

School protests send shivers down French government's spine

https://www.bbc.com/news/articles/c69wzjjxex8po
1•bookofjoe•1m ago•1 comments

What Israeli immigrants bring to America

https://www.loofwired.com/p/what-israeli-immigrants-actually
1•surprisetalk•3m ago•0 comments

We disabled PRs from external contributors on honojs/hono

https://twitter.com/yusukebe/status/2107027870019445007
1•elffjs•3m ago•0 comments

Mario and Armin on Why Pi Durable

https://twitter.com/pidotdev/status/2107033061905104941
1•tosh•3m ago•0 comments

Rigsmith, simplified 3D rigging/animating

https://github.com/tonis2/rigsmith
1•tonis2•6m ago•0 comments

Show HN: gtm, a prettier cliamp alternative with daemonized playback.

https://github.com/prjctimg/gtm
1•prjctimg•7m ago•0 comments

Categorize changes in Pull Requests into logical groups, with a summary

https://pulls.review
2•z0al•9m ago•0 comments

1970 Bhola Cyclone

https://en.wikipedia.org/wiki/1970_Bhola_cyclone
1•thunderbong•11m ago•0 comments

Uber invests in Verne, confirmed by the CEO of the Croatian robotaxi company

https://evwire.com/p/verne-ceo-confirms-uber-investment-robotaxi-europe
1•taubek•11m ago•0 comments

#227: Pivotal – Developing Perspective

https://developingperspective.com/2026/10/01/227/
1•tambourine_man•11m ago•0 comments

What If the Moon Disappeared Tonight?

https://www.youtube.com/watch?v=kStTa3zPCvE
1•mikeirvine•11m ago•0 comments

Show HN: Pure Java library for archive compression with zero dependencies

https://github.com/RealBurst/arcana
1•RealBurst•11m ago•0 comments

A health assistant that answers from Kenyan guidelines and cites them

https://tabibu.health/
1•Xandr-AJ•13m ago•0 comments

Recreating Diagrams in Pikchr #2: Dissecting the Story of Mel

https://blog.andreani.in/blog/41/
1•gandreani•14m ago•0 comments

Show HN: Native Port of Super Smash Bros. Melee

https://github.com/nikhilunni/melee
2•nikhilunni•15m ago•0 comments

How Long It Takes Google Search for Crawling, Indexing and Serving

https://www.seroundtable.com/google-crawling-indexing-serving-data-42225.html
1•thm•16m ago•0 comments

Ightline – Real-time collaborative CV annotation using CRDTs

https://sightline-crdt.netlify.app
1•joshuageoschool•17m ago•0 comments

Noah Shinn Turns Privacy Complaints into a $10B Round

https://ailately.com/articles/noah-shinn-instinct-10-billion
1•ryanedennis•19m ago•0 comments

Inference Economics: Gartner's 96 Percent

https://ailately.com/articles/inference-economics-gartner-96-percent
1•ryanedennis•19m ago•0 comments

Spending on AI Is Becoming Almost Impossible for Businesses to Budget

https://www.wsj.com/tech/personal-tech/ai-token-spending-businesses-431ee94a
2•swolpers•20m ago•0 comments

Sam Altman Sees the Future. Are We in It? (Part 2 of 2)

https://www.vanityfair.com/story/sam-altman-exclusive-interview-part-2
1•swolpers•21m ago•0 comments

Adding Reflection to C

https://www.davidpriver.com/adding-reflection-to-C.html
3•rramadass•21m ago•0 comments

My small wish for terminal programs (and other programs)

https://utcc.utoronto.ca/~cks/space/blog/sysadmin/TerminalProgramsColourModeWish
1•speckx•21m ago•0 comments

OpenAI will show visual ads in ChatGPT while you generate images

https://www.bleepingcomputer.com/news/artificial-intelligence/openai-will-show-visual-ads-in-chat...
2•thm•21m ago•1 comments

Are psychedelics anti-inflammatory? A systematic review (2027)

https://www.sciencedirect.com/science/article/pii/S0889159126007816?via%3Dihub
1•tokai•21m ago•0 comments

Computers in an Audi talk to each other

https://bogdan.nimblex.net/wired-car/
1•bogdan_r•22m ago•0 comments

I need to stop looking up reviews for everything I buy

https://nate.spot/looking-up-reviews/
2•surprisetalk•23m ago•0 comments

Show HN: Twrctrl – Browser-Based Tower and Ground ATC Simulator

https://twrctrl.com/
1•franklin0s•23m ago•0 comments