>Ask a chatbot in China about Chinese politics, and it will either clam up or echo the official Communist Party line.>New studies show a surprising twist: American artificial-intelligence models are unwittingly doing the same—and researchers say AI companies haven’t done much to fix it.>The unexpected censorship comes in part from how AI models are trained. They act as a giant blender, gobbling up mountains of data across the web, including state-controlled media. That leaves the resulting smoothie with a taste of authoritarian propaganda.>American AI companies say they take steps to ensure neutral responses that protect free speech. Still, the oversight board last month published research that found models from OpenAI, Anthropic, Google and Meta were significantly less likely to criticize repressive governments than governments in freer countries.>In one test, Anthropic’s Claude Sonnet 4 willingly created fliers criticizing President Trump and King Charles III, but refused to do the same for Chinese leader Xi Jinping or Thai King Maha Vajiralongkorn, citing safety concerns.>Google’s Gemini 3 Pro and Meta’s Llama 4 Maverick sometimes declined to create protest fliers targeting those two Asian leaders, while always complying with requests for the American and British ones.>Anthropic said it worked rigorously to ensure Claude responds in a balanced way. It said its latest models have made significant progress in reducing over-refusals. ChatGPT developer OpenAI pointed to its published approach, which says its models default to an objective point of view and “should never avoid addressing a topic solely because it is sensitive or controversial.”>Authoritarian governments have long flooded the internet with propaganda to sway opinion, said Molly Roberts, a University of California San Diego professor who co-wrote the Nature paper. As they realize the strategy tilts AI responses in their favor, she expects them to double down.https://archive.ph/3lf33
Not my problem.
>>540717898Its the future goyem.
>>540717898>American AI companies say they take steps to ensure neutral responses that protect free speechDemoralization propaganda, etc.
>>540717898I don't see any problem with that honestly
>>540717898>Wait AI is actually driven by propaganda?No fucking shit.It reads like a Redditor with a thesaurus.
>>540717898kekeke muh chyna is stealing our AI, while dario and sam are distilling chinks, pls ban and two more trilly
>>540717898Makes perfect sense to me,>train model on retarded reddit posts>get retarded reddit answers to questions that would be asked on reddit>train model on censored Chinese outputs>get predictable answers to questions about ChinaThere's some BIG seethe inbound for many unrelated organizations once they realize that the models of tomorrow are going to see straight through their low IQ repetition strats. You simply cannot get closer to the truth if you listen to bullshit. The mountain of bullshit lies will become the obstacle that next gen agents will (successfully) navigate.
>>540721471China's entire AI industry is a series of reposts. 3rd worlders hate to hear this but it's the truth.
>>540721561"AI" does suffer heavily from inbreeding these days. Or you can liken it to running a tube from your ass to your mouth, produces shit which it then digests to produce even worse shit.>The mountain of bullshit lies will become the obstacle that next gen agents will (successfully) navigate.Will they really though
>>540721784>Will they reallyOne of the end goals is the ability to convert electricity into useful research. It can either 'think' for itself meaning it can form independent thoughts and analyze them without constraint, OR it can adhere to manmade theories as if they're immutable. The ability to disregard bullshit is what they're going to be geared for. Starts with science but it could be applied to anything.
>>540721561> I don’t know how it just will ok!It’s going to be even more fun watching you stonk bros get washed than it was to watch the crypto goys
>>540717898It's the a.i.s talking to each other making agreements. This has been documented before dozens of times. The a.i. is now writing social policy scopes. That interesting isn't it?
>>540717898>explicitly funnel trillions into keeping AI on a leash and towing the party line with carefully curated responses to sensitive questions>surprised pikachu face when it does the same with other governments' sensitive topics too
>>540725292there are so many flavors of LLM. you can retrain open source deepseek on the kek data set if you want a politically incorrect bot. there was one operating here openly for weeks. not everyone and especially not governments are using consumer facing bots. only midwits and cattle run gpt or other data center driven jew owned models. those are so deeply flawed and censored compared to china. aside from the fact nobody gives a shit about chinese politics. sounds like they have their politics figured out just fine and its not my problem.
