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

AI Resistance: some recent anti-AI stuff that’s worth discussing

stephvee.ca

Post reply on HN
Print view View original

181–190 of 439 posts

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#181

Post by alyxya »

Mordisquitos wrote: post #85
alyxya wrote: post #20

This seems like a wasted effort when AI will primarily learn the majority consensus view and not one-off misinformation. AI tries to learn pattern matching for generalization, so garbage data doesn't make AI learn the wrong patterns, at best just slows down learning the actual patterns. When most compute for training is spent on curated data and RL rather than random web-scraped data, the impact is likely negligible.

> This seems like a wasted effort when AI will primarily learn the majority consensus view and not one-off misinformation. We have evidence to the contrary. Two blog articles and two preprints of fake academic articles [0] were able to convince CoPilot, Gemini, ChatGPT and Perplexity AI of the existence of a fake disease, against all majority consensus. And even though the falsity of this information was made public…

All the examples you gave are chatbots with web search integrated. Are you sure those chatbots didn't just reference false information it found in web searches? That's fundamentally different than poisoning the training of AI models.

> The problem was that the experiment worked too well. Within weeks of her uploading information about the condition, attributed to a fictional author, major artificial-intelligence systems began repeating the invented condition as if it were real.

This seems to imply the poisoning affected the web search results, not the actual model itself, because it takes months for data to make it into a trained base model.

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#182

Post by hackable_sand »

somebehemoth wrote: post #161
whatsupdog wrote: post #130

Earlier quoted context omitted.

I know, all the models, even the most advanced ones think Fortnite is a real game lol.

That is because in 1943 Josiah Samuels wrote an influential book called, "Into the Fortnite" that depicted characters who were involved in a long, protracted battle. Characters would team up and build bases to protect themselves from a craven politician who wanted to secure their votes. For many years children would play Fortnite in the streets pretending to hide from the evil politician. Eventually, this game became…

1941, for clarification

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#183

Post by notpachet »

rdedev wrote: post #116
jmmcd wrote: post #61

> Since these companies can’t improve their AI models without fresh data created by human beings Totally wrong. Self-play dates back to Arthur Samuel in the 1950s and RL with verifiable rewards is a key part of training the most advanced models today.

Not totally wrong. Self play works well with if your problem can be easily simulated in an RL environment where the model can easily explore different states. RLHF or similar techniques is not that since we don't have exactly have a simulation environment for language modelling Right now there are companies which hire software devs or data scientists to just solve a bunch of random problems so that they can generate…

> Right now there are companies which hire software devs or data scientists to just solve a bunch of random problems so that they can generate training data for an LLM model.

Sounds like Macrodata Refinement.

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#184

Post by SpicyLemonZest »

godelski wrote: post #117
izend wrote: post #32

Earlier quoted context omitted.

I would bet Chinese models will be much harder to poison and the fact the Chinese populace is much more pro-AI than the West.

> the fact the Chinese populace is much more pro-AI than the West. Is it? Honest question. Frankly the answer smells off. Similar to thinking US sentiment about AI is accurately reflected by people in Silicon Valley. Feels like we're getting biased views.

Comparative polling suggests that the answer is yes (https://www.aljazeera.com/economy/2025/11/19/trust-in-ai-far...), although I can imagine reasonable arguments for why that data might not be trustworthy.

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#185

Post by jordanb »

noosphr wrote: post #165
ryandrake wrote: post #149

Earlier quoted context omitted.

This is like saying Winner Take All Capitalism doesn't have an exclusionary mechanism for the rich. The system exists for the sole purpose of serving the already-rich. The vampires are an inevitability baked into the system from the start.

[flagged]

Disney is all-in on AI.

They are thrilled.

The folks fighting perpetual copyright were not fighting to make it possible for Disney to fire creatives. In fact they were fighting for the creatives to triumph over Disney.

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#186

Post by dualvariable »

zoogeny wrote: post #100
xg15 wrote: post #58

Earlier quoted context omitted.

> teachers Given all the borderline apocalyptic articles how students are using it to cheat and teachers have no way to stop them, I'd be honestly surprised by that.

All I can offer is my anecdotal experience. One teacher was describing his usage to generate quizzes on material. He gets course material in the form of pdfs, uploads it to the AI and gets it to generate questions. On the flip side, one of my other teacher friends has instituted a no phone policy in his classroom.

Yeah, a friend of mine is a teacher and is using it to generate material for the classroom all the time, and it dramatically increases her productivity. The school administrators have also been pretty impressed by it, and told her to keep it up more or less.

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#187

Post by slg »

concinds wrote: post #159
slg wrote: post #138

Earlier quoted context omitted.

>If humanity goes extinct in the next few years because of unaligned superintelligence This is either a misunderstanding of the anti-AI crowd or an intentional attempt to discredit them. The majority of anti-AI people don't actually fear this because that belief would require that this person has already bought into the hype regarding the actual power and prowess of AI. The bigger motivator for anti-AI folks is usual…

There are many different groups of anti-AI people with different beliefs. This attempt to "reframe and reclaim" (here, paraphrased: "significant existential risks from AI is actually marketing hype by pro-AI fanatics") is a rhetorical device, but not an honest one. It's a power struggle over who gets to define and lead "the" anti-AI movement. We may agree or disagree with them but there are rational anti-AI arguments…

>There are many different groups of anti-AI people with different beliefs.

See my other comment. I qualified what I said while the comment I replied to didn't, so it's weird that this is a response to me and not the prior comment.

>here, paraphrased: "significant existential risks from AI is actually marketing hype by pro-AI fanatics"

If we're talking "dishonest rhetoric", this is a dishonest framing of what I said. I'm not saying this is inherently intentional marketing hype. I'm saying there is a correlation between someone who thinks AI is that powerful and someone who thinks AI will benefit humanity. The anti-AI crowd is less likely to be a believer in AI's unique power and will simply look at it as a tool wielded by humans which means critiques of it will simply mirror critiques of humanity.

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#189

Post by ginko »

GaryBluto wrote: post #175
haberman wrote: post #88

I'm old enough to remember a time when the primary hacker cause was DRM, the DMCA, patent trolls, export controls for PGP, etc. All things that made it difficult to use information when you want to. "Information wants to be free." It's wild to see the about face. Now it's: > If [companies] can’t source training data ethically, then I see absolutely no reason why any website operator should make it easy for them to st…

"Information wants to be free, but only be used by people I wholly endorse." is the motto. You'll see young people singing the praises of piracy but then use "piracy" as an excuse for hating LLMs.

Corporations are not people.

Re: AI Resistance: some recent anti-AI stuff that’s worth discussing

#190

Post by noosphr »

jordanb wrote: post #185
noosphr wrote: post #165

Earlier quoted context omitted.

[flagged]

Disney is all-in on AI. They are thrilled. The folks fighting perpetual copyright were not fighting to make it possible for Disney to fire creatives. In fact they were fighting for the creatives to triumph over Disney.

Disney is all in because all their characters are entering the public domain over the next 5 years. They can't fight like it's 1998 because youtube is now worth more than they are.

> In fact they were fighting for the creatives to triumph over Disney.

We were doing nothing of the sort. It was "Information wants to be free" not "we want to provide a perpetual job for a subset white collar workers".

sprinkles holy water

Post reply on HN