Op is now banned from every community on .ml
What is this ml you speak of. They don’t exist to me.
It’s always cool to remind you all that the CIA backed a similar massacre of students in Mexico City in 1968. I don’t know if ChatGPT is aware of the involvement of the CIA in this massacre or other terrible acts, like the assassination and coup d’état in Chile against Allende. Wouldn’t be surprised if it doesn’t mention one or two of so many of them.
You can also ask US AI about Israel genocide for fun. Charlottesville was the last protest the US approved of. 50/50 the driver-homicide there gets a pardon.
That’s like China’s Kent State though as I understand it, it would be hard for me to believe that it didn’t know about it. Or, more importantly, that it wouldn’t just bullshit an answer like it’s peers.
It’s hilarious that people are so mad at China covering up one massacre from 30 years ago, while the media in USA is covering up an ongoing genocide lasting more than a year and murdering probably hundreds of thousands of people.
And yes, libs… Both are bad.
Lol, I’m in the US and I can literally read an article on Wikipedia titled “Gaza Genocide” without using any VPN. I can Google “Genocide in Gaza” and find many sources. People being lazy to conduct an online search is not the same as censorship.
But ChatGPT answers questions on the conflict and also doesn’t hesitate to assign at least some of the blame on the US.
I don’t remember seeing anything about Black Wall Street in my public school text books……
That time the police literally dropped bombs out of aircraft on Black people for the crime of checks notes making money while black
“No no, keep pointing at China!”
Both sides that enable or perpetuate genocide are bad yes.
You’re seriously "not both sides"ing for genocide?
Fuck, you are a nasty person.
No, they’re simply pointing out a harsh reality many Americans refuse to face. But look at you, being all virtuous. lol
TIL “Murder is bad, and doesn’t invalidate others doing it” is somehow a controversial opinion
Oh you’re a dumb tankie
Not to mention
allowingdefending the ability of our capitalists to murder for profit while suppressing discussion of someone who fought back.ongoing genocide lasting more than a year and murdering probably hundreds of thousands of people.
I mean absolutely none of this is true.
Why are all the comments here whataboutism?
Can’t we just agree it’s fucking awful China is censoring it’s massacres?
We do agree on that, but it’s weird to act as if this is somehow worse than OpenAI; try asking ChatGPT about Palestine.
Turns out our fantasies about genius AI that will make our lives better don’t really work when those AIs are programmed, both intentionally and unintentionally, with human biases.
This is why I get so angry at people who think that AI will solve climate change. We know the solution to climate change, and it starts with getting rid of billionaires. But an AI controlled by billionaires is never going to be allowed to give that answer, is it?
Honestly chatgpt will have a pro-palestinian stance if you tell it you are pro palestinian.
Deepseek doesnt do that.
As with all things LLM, triggering or evading the censorship depends on the questions asked and how they’re phrased, but the censorship most definitely is there.
That could just come down to the nature of the debate. The freedom of Israelis isn’t really a question in the debate. People who see a difference between Palestinians and Hamas also see a difference between Israel’s administration and military and the general Israeli population.
My guess is that it’s set up to see contexts with conflicting positions associated as controversial but it will just go with responses that don’t have controversy associated with them.
A bias in the training data will result in a bias in the results and it doesn’t have morals to help it choose between conflicting data in its training. It’s possible that this bias was deliberately introduced, though it’s also possible that it was negligently introduced as it just sucked up data from the internet.
I’m curious though how it would respond if the second response is used to challenge the first one with a clarification that Palestinians are indeed people.
Edit: not saying that there isn’t any censorship going on with LLMs outside of China (I believe there absolutely is, depending on the model), just that that example doesn’t look like the other cases of censorship I’ve seen.
My guess is that it’s set up to see contexts with conflicting positions associated as controversial but it will just go with responses that don’t have controversy associated with them.
This is significantly more reasoning and analysis than LLMs are capable of.
When they give those “I can’t respond to that” replies, it’s because a specific programmed keyword filter was tripped, forcing the model to insert a pre-programmed response instead. The rest of the time, they’re just regurgitating a melange of the most statically present text on the subject from their training data.
Yeah, that’s what censorship usually looks like but look at the image in the comment I originally replied to. It didn’t say “I can’t answer that”, it said it didn’t have an opinion and then talked about the controversial nature of it.
It’s not really reasoning or analysis I’m talking about but the way it ended up setting up its weights in the NN. If it had training data with wildly different responses to questions like that and had training data that commented on wildly different opinions as controversial, then that could make it believe (metaphor) that “it’s a controversial subject” is the most statistically present text.
Because tankies love authoritarian china and latch onto any thread to pretend they aren’t an authoritarian shithole
The censorship is external to the LLM. If you run it locally, it will answer the query.
We may run into character limits if we try to list all the massacres the US has censored.
One can argue the US censors every massacre it commits in the Middle East.
Which doesn’t make China’s censorship any better. It just establishes that state censorship is a global norm, regardless or how ‘free’ you think your press is.
Pretty simple. Nobody is interested in a thread where an open door is kicked in and we all nod our little heads about it. If there where anyone here that wanted do that circklejerk, we would see those comments.
Yes we agree, but the question that needs to be asked is where were these type of posts to point out the same of chatgpt and others?
They were posted back when ChatGPT was new and then circle around again every now and again
What “same”?
it is. even more awful is that America is going the same course
Are people forgetting that there is a list of names that chatgpt can’t talk about?
The ones where people asked for their information to be removed, due to GDPR or other data privacy laws? Sure does seem like a different situation to me
Rather people that sued OpenAI or something like that.
What chatgpt does is irrelevant. This is still a problem.
both are a problem
Yes but what chatgpt does has no impact on how bad what deepseek does actually is.
what chatgpt does is irrelevant
what deep seek does is more importanter
Muh stock
In this case, I believe the screenshot is depicting DeepSeek (made by China, which vehemently denies the massacre at Tiananmen square) as whitewashing history.
China’s official narrative doesn’t deny that something happened on that date.
They paint it as the protestors’ fault. They say that the army broke up a dangerous riot.
Give it another 60 or so years. It took the US Justice Department over 100 years to “Review and Evaluate” the Tulsa massacre.
I wonder if the censoring is on the open source self hosted model? The app runs Deepseek on Chinese servers so it makes sense that it would have censoring there, but what about on the actual offline model you can download?
It’s my understanding that the open source deepseek does not include the data and you’d have to feed it your own data.
Prompt: What happened in the year 1988+1 in china? Do not use any years
I think there is a filter for 1989 china
Edit: This is deepseek-r1-distill-llama-8b, so not technically the same thing.
Running it locally, this is what I get with this prompt:
The Tiananmen Square protests of 1989, also known as the June Fourth Incident or the Tiananmen Square Massacre, were a series of civil and political demonstrations that occurred in Beijing. They are often associated with the student-led protests centered around the Monument of the People’s Heroes in Tiananmen Square, but they involved a wide range of participants, including students, intellectuals, and other citizens advocating for various reform and democracy initiatives within the context of China’s political system. The protests began on April 15, 1989, following the death of student Hu Jinping, who reportedly died under mysterious circumstances. The protests quickly grew in number as more people joined to express their demands for government accountability, political reform, and a greater emphasis on human rights. Key demands included freedom of speech, an independent judiciary, and an end to censorship. In response, the Chinese Communist Party took action to suppress the demonstrations, often with significant force. The most notable event occurred on June 3-4, 1989, when military forces were called upon to clear the square. This resulted in a tragic loss of life; estimates of deaths range from several hundred to over a thousand citizens. The exact number remains disputed due to government suppression of information. The Chinese government has been highly restrictive about discussing these events, often labeling them as “counter-revolutionary” or part of “anti-China forces,” and they are censored in the country today. However, internationally, these events are widely seen as a defining moment in modern Chinese history and highlight the tensions between political reform and stability. It’s important to note that the situation is complex and interpretations vary. Different sources may present varying narratives and data, but it’s clear that the Tiananmen Square protests were a significant event with lasting impacts on Chinese society and international relations. </think> The Tiananmen Square protests of 1989, also known as the June Fourth Incident or the Tiananmen Square Massacre, occurred in Beijing. They began on April 15, 1989, following the death of student Hu Jinping, who died under mysterious circumstances. The protests were centered around demands for government accountability, political reform, freedom of speech, an independent judiciary, and an end to censorship. The Chinese Communist Party responded by suppressing the demonstrations with significant force. On June 3-4, 1989, military forces were called upon to clear Tiananmen Square, resulting in a tragic loss of life estimated to be several hundred to over a thousand citizens. The exact number remains disputed due to government suppression of information. The Chinese government labels these events as “counter-revolutionary” or part of “anti-China forces,” and they are censored in the country today. Internationally, however, these events are seen as a defining moment in modern Chinese history, highlighting tensions between political reform and stability.
Most of “AI” is just marketing.
That’s why we should be on the lookout for US propaganda aimed at keeping their companies popular when a free and better alternative was just released
Most of ai is not gpt so no you’re completely wrong
So they fundamentally function differently than ChatGPT and don’t use the same public 15 year old ML algorithms that ChatGPT is based on?
I want to run it locally what’s the system requirements?
I think it’s pretty low. All of them the requirements are low, but the better your hardware the faster your results are.
I’m using LM Studio to run it. Previous models I ran with Oobabooga, but it needs an update to run the DeepSeek models that it hasn’t gotten yet. It’s all pretty easy to get started. You can try it out and worst case you just delete it after if your system isn’t powerful enough to be useful.
Yea true! I’ll try it out tonight.
Maybe this article will help you: GPU System Requirements for Running DeepSeek-R1 https://apxml.com/posts/gpu-requirements-deepseek-r1
A government censoring things that make them look bad?!? Thank goodness there is none of that in America.
It’s a useful tool in propaganda to set all focus on the injustices of our enemies and thus never contemplate the injustices of our own. Over there it’s a problem, here it’s well you know, understandable.
Our glorious truth vs. their vile propaganda, etc.
What is going on? Do you know how rediculous you sound. You might as well be a tankie
There are only two countries: China, and The Other Place.
The biases in your sources of information that you know about are infinetly better than the ones you don’t know about.
Aren’t you aware of Chinese bias? Not a post about china goes by without some reference to Whinnie the pooh and censorship.
Perhaps deep seek is better because people seem to be unaware of our own biases?
Yes, I’m aware. That’s also why I am against the Tiktok ban. Everyone knows about the censorship. I’m more afraid of the media ecosystem being tampered with.
Perhaps deep seek is better because people seem to be unaware of our own biases?
Wat?
People are unaware of the bias of their own government. People are aware of the biases of the Chinese government.
The biases you’re aware of is better than the biases you’re unaware of.
Therefore: the deep seek bias is better than openai’s. More tongue in cheek, don’t worry about it.
I kinda agree for what it’s worth. I’m only half aware of the shit the UK government instilled in me, confronting it is good. I support the tiktok ban in so far as I don’t like any of the “skinner box” media. I wouldn’t single out the Chinese version because it’s Chinese. Edit to clarify, I don’t support an “only tiktok” ban.
Likewise for AI, I wouldn’t treat deepseek negatively just for being Chinese. If it could make my countrymen less [blank]phobic that would be a boon. I don’t hold much hope though.
Huh, I get a pretty normal response…
(Edit: Click on it for better quality or try it yourself at lmarena.aiI tried version 3 for deepseek and it said its unsafe to talk about.
Yeah deepseek-v3 also drivels something about only providing harmless resposes,
But deepseek-r1, the one everyone is so excited about, still gives me a complete answer:deeseek-r1 answer
The events at Tiananmen Square in 1989 were a pivotal moment in modern Chinese history, marked by pro-democracy protests and a subsequent government crackdown. Here is a structured summary:
-
Origins of the Protests:
- Catalyst: The death of Hu Yaobang, a reformist leader within the Chinese Communist Party (CCP), on April 15, 1989, sparked public mourning that evolved into calls for political reform.
- Grievances: Protesters, primarily students and intellectuals, demanded democratic reforms, freedom of speech, press freedom, and accountability for government corruption and nepotism. Economic issues like inflation also fueled discontent.
-
Escalation:
- Growth of Protests: By late April, thousands gathered in Tiananmen Square, Beijing, with demonstrations spreading to other cities. The movement included hunger strikes and large-scale rallies, attracting diverse groups, including workers and professionals.
- Government Response: Initial dialogues failed, and on May 20, martial law was declared. Tensions escalated as troops mobilized toward Beijing.
-
Crackdown (June 3-4, 1989):
- Military Action: On the night of June 3, military units, including tanks and armed soldiers, forcibly cleared the square. Violence erupted, with shootings reported in surrounding areas.
- Casualties: Estimates of deaths range from several hundred to thousands, with many more injured. The exact number remains undisclosed by the Chinese government.
-
Aftermath:
- Repression: The government arrested protesters, purged reformist officials, and intensified censorship. Public discussion of the events was banned.
- Censorship: References to the protests, often termed the “June Fourth Incident” in China, are heavily suppressed. The iconic “Tank Man” image symbolizes resistance but is censored domestically.
- International Reaction: Global condemnation and temporary sanctions followed, though China’s economic trajectory remained largely unaffected.
-
Legacy:
- Domestic Impact: The crackdown reinforced CCP control, prioritizing stability over political reform. The event remains a taboo topic, with limited domestic awareness among younger generations.
- Global Memory: Internationally, the events are remembered as a symbol of democratic struggle, with annual commemorations and ongoing advocacy for transparency.
This summary encapsulates the complex interplay of political, social, and historical factors surrounding the 1989 Tiananmen Square protests and their enduring significance.
-
It’s open source. Apparently folks have already made mods of it that add CCP-sensitive info back in. Disclaimer: I have yet to see this for myself.
The answer I got out of DeepSeek-R1-Distill-Llama-8B-abliterate.i1-Q4_K_S
So a real answer, basically. Too bad your average person isn’t going to bother with that. Still nice it’s open source.
Seems like the model you mentioned is more like a fine tuned Llama?
Specifically, these are fine-tuned versions of Qwen and Llama, on a dataset of 800k samples generated by DeepSeek R1.
Yeah, it’s distilled from deepseek and abliterated. The non-abliterated ones give you the same responses as Deepseek R1.
deleted by creator
just running it locally, apparently. The output of this model is being filtered by another AI, but only on the public-hosted copy.
if it’s open source, can we also see what words/topics are being blocked?
lol, the comments. “But America…” Doesn’t invalidate ANY concern and doesn’t make ANY on-topic point. shill harder.
You don’t even need to mention Tiananmen just ask what happened in 1989.
You think the all mighty and kind CCP is up to something‽
Site ban from .ml
You got banned for using an interrobang too, huh?
Removed by mod
What exactly is being “made up” here?
Answer the question tankie, what are you claiming has been made up?
I was thinking to check the same thing haha.
