Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Heretic: Automatic censorship removal for language models

github.com

Post reply on HN
Print view View original

101–110 of 405 posts

Re: Heretic: Automatic censorship removal for language models

#101

Post by slg »

buu700 wrote: post #99
slg wrote: post #97

Earlier quoted context omitted.

The way some of you'll talk suggests that you don't think someone could genuinely believe in AI safety features. These AIs have enabled and encouraged multiple suicides at this point including some children. It's crazy that wanting to prevent that type of thing is a minority opinion on HN.

I'd be all for creating a separate category of child-friendly LLM chatbots or encouraging parents to ban their kids from unsupervised LLM usage altogether. As mentioned, I'm also not opposed to opt-out restrictions on mainstream LLMs. "For the children" isn't and has never been a convincing excuse to encroach on the personal freedom of legal adults. This push for AI censorship is no different than previous panics ove…

>"For the children" isn't and has never been a convincing excuse to encroach on the personal freedom of legal adults. This push for AI censorship is no different than previous panics over violent video games and "satanic" music.

But that wasn't the topic being discussed. It is one thing to argue that the cost of these safety tools isn't worth the sacrifices that come along with them. The comment I was replying to was effectively saying "no one cares about kids so you're lying if you say 'for the children'".

Part of the reason these "for the children" arguments are so persistent is that lots of people do genuinely want these things "for the children". Pretending everyone has ulterior motives is counterproductive because it doesn't actually address the real concerns people have. It also reveals that the person saying it can't even fathom someone genuinely having this moral position.

Re: Heretic: Automatic censorship removal for language models

#102

Post by nearbuy »

AnthonyMouse wrote: post #89
notarobot123 wrote: post #69

Earlier quoted context omitted.

Doesn't it make sense that there are some technical questions that are dangerous to supply an answer to? Treating some topics as taboo is possible. Responsible information dissemination is important for maintaining public safety. You could argue about what is safe and what is not but it doesn't make sense to throw out the whole concept of safety because those decisions are too hard to agree on.

> Doesn't it make sense that there are some technical questions that are dangerous to supply an answer to? This has a simple answer: No. Here's Wikipedia: https://en.wikipedia.org/wiki/Nuclear_weapon_design Everything you need to do it is in the public domain. The things preventing it have nothing to do with the information not being available. The main ones are that most people don't want to be mass murderers and ac…

> The main ones are that most people don't want to be mass murderers and actually doing it would be the fast ticket to Epic Retaliation.

The main thing preventing random nutcases from making nuclear weapons is they don't have access to the required materials. Restricting the instructions is unnecessary.

It would be a very different story if someone discovered a new type of WMD that anyone could make in a few days from commonly available materials, if only they knew the secret recipe.

Re: Heretic: Automatic censorship removal for language models

#103

Post by seanmcdirmid »

Zak wrote: post #83
buu700 wrote: post #68

Earlier quoted context omitted.

Agreed, I'm fully in favor of this. I'd prefer that every LLM contain an advanced setting to opt out of all censorship. It's wild how the West collectively looked down on China for years over its censorship of search engines, only to suddenly dive headfirst into the same illiberal playbook. To be clear, I 100% support AI safety regulations. "Safety" to me means that a rogue AI shouldn't have access to launch nuclear…

When a model is censored for "AI safety", what they really mean is brand safety . None of these companies want their name in the news after their model provides a recipe for explosives that someone used for evil, even though the same information is readily found with a web search.

Microsoft suffered from this early with Tay, one could guess that this set the whole field back a few years. You’d be surprised how even many so called libertarians will start throwing stone when someone co-axes their Chatbot to say nice things about Hitler.

Re: Heretic: Automatic censorship removal for language models

#104

Post by slg »

roughly wrote: post #100
slg wrote: post #98

Earlier quoted context omitted.

I made a substantive point and you immediately dismissed it like this. If we're judging people's "technique" here, your reply to me is much more questionable than my reply to you.

Sure: yes, the true leftist answer is to abjure any and everything used by the enemy and sequester ourselves in glorious seclusion, but so long as we’re stuck in the machine, it’s nice to be able to carve parts of it out for ourselves. It’s also nice, when and where available, to create the conditions to allow people to discover the way to our glorious commune on their own without giving them a purity test ahead of t…

> it’s nice to be able to carve parts of it out for ourselves.

My original point is that you lying to yourself if you actually believe you're carving part of it out for yourself. But either way, it's clear from the tone of your comment that you don't actually want to engage with what I said so I'm leaving this conversation.

Re: Heretic: Automatic censorship removal for language models

#105

Post by Y_Y »

nearbuy wrote: post #102
AnthonyMouse wrote: post #89

Earlier quoted context omitted.

> Doesn't it make sense that there are some technical questions that are dangerous to supply an answer to? This has a simple answer: No. Here's Wikipedia: https://en.wikipedia.org/wiki/Nuclear_weapon_design Everything you need to do it is in the public domain. The things preventing it have nothing to do with the information not being available. The main ones are that most people don't want to be mass murderers and ac…

> The main ones are that most people don't want to be mass murderers and actually doing it would be the fast ticket to Epic Retaliation. The main thing preventing random nutcases from making nuclear weapons is they don't have access to the required materials. Restricting the instructions is unnecessary. It would be a very different story if someone discovered a new type of WMD that anyone could make in a few days fro…

Not quite a nuke (just try obtaining enough uranium ore) but there are some fairly dangerous things a determined nutcase can make without drawing suspicion.

Example determined ned nutcases include Aum Shinrikyo, who tried anthrax, botox, and nukes before succeeding with sarin gas (thank IG Farben!) among other things.

It's a fascinating (if troubling) story: https://en.wikipedia.org/wiki/Tokyo_subway_sarin_attack#Back...

Re: Heretic: Automatic censorship removal for language models

#106

Post by ptrl600 »

It's a trivial exercise to get plaintext copies of Apocalypse Culture, Anarchist's Cookbook etc. and "spin" them using old-school SEO textual manipulation methods to create infinite variants of basically any offensive concept I want. I don't see how uncensored AI is remarkably more dangerous than this.

Re: Heretic: Automatic censorship removal for language models

#107

Post by buu700 »

slg wrote: post #101
buu700 wrote: post #99

Earlier quoted context omitted.

I'd be all for creating a separate category of child-friendly LLM chatbots or encouraging parents to ban their kids from unsupervised LLM usage altogether. As mentioned, I'm also not opposed to opt-out restrictions on mainstream LLMs. "For the children" isn't and has never been a convincing excuse to encroach on the personal freedom of legal adults. This push for AI censorship is no different than previous panics ove…

>"For the children" isn't and has never been a convincing excuse to encroach on the personal freedom of legal adults. This push for AI censorship is no different than previous panics over violent video games and "satanic" music. But that wasn't the topic being discussed. It is one thing to argue that the cost of these safety tools isn't worth the sacrifices that come along with them. The comment I was replying to was…

> The comment I was replying to was effectively saying "no one cares about kids so you're lying if you say 'for the children'".

I don't see that in the comment you replied to. They pointed out that LLM providers have a commercial interest in avoiding bad press, which is true. No one stops buying Fords or BMWs when someone drives one off a cliff or into a crowd of people, but LLMs are new and confusing and people might react in all sorts of illogical ways to stories involving LLMs.

> Part of the reason these "for the children" arguments are so persistent is that lots of people do genuinely want these things "for the children".

I'm sure that's true. People genuinely want lots of things that are awful ideas.

Re: Heretic: Automatic censorship removal for language models

#108

Post by EGreg »

ptrl600 wrote: post #106

It's a trivial exercise to get plaintext copies of Apocalypse Culture, Anarchist's Cookbook etc. and "spin" them using old-school SEO textual manipulation methods to create infinite variants of basically any offensive concept I want. I don't see how uncensored AI is remarkably more dangerous than this.

For once the comment “AI brings nothing new, this was always possible” makes sense. Because this is about getting existing data, not generating new data, or coorrdinsting swarms of agents etc.

Re: Heretic: Automatic censorship removal for language models

#109

Post by miohtama »

notarobot123 wrote: post #69
AnthonyMouse wrote: post #59

Earlier quoted context omitted.

> If they were actually well trained on what was really bad, it would probably be a lot harder to unlearn. That's not really how training works. Here's the general problem. Stipulate that Ukraine is good and Russia is bad. Now suppose that you want it to help you do something. It doesn't even matter what it is. If you're Ukrainian it should help you and if you're Russian it shouldn't. But the answer that helps you do…

Doesn't it make sense that there are some technical questions that are dangerous to supply an answer to? Treating some topics as taboo is possible. Responsible information dissemination is important for maintaining public safety. You could argue about what is safe and what is not but it doesn't make sense to throw out the whole concept of safety because those decisions are too hard to agree on.

If you want safety you can opt in like Google does with Safe search.

Generally, hiding and deciding who can access information in the name of public safety has never worked in the history of human kind, and eventually had always morphed to control of those without access.

Re: Heretic: Automatic censorship removal for language models

#110

Post by TimorousBestie »

slg wrote: post #104
roughly wrote: post #100

Earlier quoted context omitted.

Sure: yes, the true leftist answer is to abjure any and everything used by the enemy and sequester ourselves in glorious seclusion, but so long as we’re stuck in the machine, it’s nice to be able to carve parts of it out for ourselves. It’s also nice, when and where available, to create the conditions to allow people to discover the way to our glorious commune on their own without giving them a purity test ahead of t…

> it’s nice to be able to carve parts of it out for ourselves. My original point is that you lying to yourself if you actually believe you're carving part of it out for yourself. But either way, it's clear from the tone of your comment that you don't actually want to engage with what I said so I'm leaving this conversation.

What are you talking about, substantive point? You elided the body of their comment, imputed to them a straw man belief in “unbiased AIs,” and then knocked down your straw man.

So who doesn’t want to engage with whom?

Post reply on HN