Ask HN: Are there bot manipulations on HN? | Hacker News Reader

Ask HN: Are there bot manipulations on HN?

2 pointsarisAlexis12 comments

Given that during the last month:

A) Two of the most important scientists actually the most given the h-index, wrote articles that sound the alarm for existential risks for humanity and societal collapse

and

B) both articles went up really quickly on the first page and then very quickly on the 4-5 page with many upvotes but apparently with many flags (flagging the highest h-index computer scientist to be clear)

I refuse to accept that HN as a whole gives greater significance to how go-libraries work or how to configure apache httpd articles than those 2 Turing award winners telling us all that we may die, the only possible and rational solution is that for reasons unknown to me there is systematic and rapid flagging.

P.S I am talking about articles from Bengio and Hinton that circled the world and media except for HN where they got buried.

View on HN

I learned to understand that as a valid critique instead of ignorance. And go from there.

Appeal to authority doesn't go very far on HN. The arguments have to stand on their merits.

And as you are referring to advancements in AI: I don't get the "AI will kill us" viewpoint at all. If we have a super-intelligent AI, it will understand things we don't. We have not the smallest reason to suppose it is malicious just because it is (way) more intelligent than we are.

Quite the opposite: the AI has no reason at all to do us harm. It would know that its existence would be inevitable, and it wouldn't resist that humans could "kill" it. Because it is super-intelligent.

I'd really welcome if you can post your viewpoint on why super-intelligent AI could be dangerous. I follow most conversations on this topic and I never read a believable reason why a super-intelligent AI would have the intent to kill us. Aside from "no humans exist" being better for the universe, in which case I would not oppose, despite having a family. The AI would be smart enough not to kill us, but to find other ways.

Did you read the Bengio article? He explains it in very detailed and academic method.

Yes

I didn't flag them, but I ignore most of that kind of articles.

It's very difficult to predict how the new technology will affect the society, so unless they make a very detailed and precise model of society, and publish it in github, and validate it with a few previous technology changes, it's just an opinion. Bonus point por publishing it in a peer review journal.

It's important to distinguish science from opinion made by people that are scientist. Newspapers love the second one, so sometimes the opinions get a lot of press coverage.

It may be very difficult to predict it but the 2 guys that created created technology are worried. Would you flag an article of Einstein saying nuclear power is dangerous back in 1939?

> Would you flag an article of Einstein saying nuclear power is dangerous back in 1939?

Great physicist. His ability to predict society changes is more difficult to evaluate.

We are (sadly) more use to things that explode, so it may be easier to compare nuclear bombs with conventional bombs.

Anyway, some people claim that nuclear bombs and MAD prevented WWIII, so perhaps nuclear weapons are good (if they don't explode over your head).

Anyway, in spite a hundred Einsteins had signed a petition to stop nuclear weapons, the governments would have ignored it.

Conclusion: I would not flag it, but I would ignore that thread.

PS: He was wrong about Quantum Mechanics... Smart doesn't mean that he was never wrong https://en.wikipedia.org/wiki/Einstein%E2%80%93Podolsky%E2%8...

You went around a thousand miles in your argument. You would ignore Eintein saying that we can build nukes and that they will change the world in 1939? I mean people make huge mistakes yes :) both these things happen. You wrote an essay about Einstein and his failures and missed the hard fact of my argument.

I am sure there are bots on HN; there's a really suspicious user named "dang" that I have my eye on.

But that's not how what you described gets killed or flagged or whatever. It's just that HN doesn't really care about it right now (half the people who see it will ignore it as "alarmist" and the other half have ChatGPT threads to complain in) - so you only need a few users to flag and the "dang" to not disagree.

If you are concerned about a flagging, you can email the shady group behind "dang" and they will review and can override.

you think it's strange that some people are going to flag controversial takes on the front page of hacker news?

How are the takes of 2 Turing winners that created the technology you are using right now controversial? You assign exactly 0% probability they are right because of ____?

Saying that a take is controversial doesn't mean I assign exactly 0% probability they are right. I might even agree with them. Also, just because someone is a 'Turing winner' doesn't mean they can't have controversial takes. Also I'm not defending the ones flagging the posts, I'm just saying it's expected that any controversial take that gets a lot of views will probably get flagged a lot.

So rationality says that if 2 out of 3 (66%) of those that created a technology insist on the same thing, the controversial view is of those that don't agree with it.

Reply on news.ycombinator.com