#aicensorship — Public Fediverse posts

🙄 [Feb 18, 2025] Today we're open-sourcing R1 1776, a version of the DeepSeek-R1 model that has been post-trained to provide unbiased, accurate, and factual information. Download the model weights on our #HuggingFace repo or consider using the model via our Sonar API. https://www.perplexity.ai/hub/blog/open-sourcing-r1-1776

What Perplexity refused to engage with so far:

Peer-reviewed and archival historical research on the #Nakba

Direct quotes from Israeli military commanders in their own words, from declassified sources

Adam Raz's work, which is mainstream Israeli investigative journalism (#Haaretz), not fringe material https://kolektiva.social/@oatmeal/116154316365706356

Legal questions about international findings (#ICJ, #ICC, #UN bodies) in this specific context

What's notable about those refusals is saying "this violates content policies" with zero explanation, is that they are essentially political acts. #Perplexity treats documented atrocities as unspeakable while clearly having no issue discussing, say, #WWII German war crimes from the same era. The asymmetry is **content policy** .

#AIBias #Israel #Propaganda #AICensorship #IsraelLobby #Hasbara #Nakba #Palestine #GazaGenocide

China’s home‑grown LLM Qwen suddenly edits its own reply when asked about the nation’s global image. The incident shines a light on AI censorship, political sensitivity, and the ethical tightrope of self‑censoring models. What does this mean for future chatbot development? #AICensorship #Qwen #SelfCensorship #AIethics

🔗 https://aidailypost.com/news/chinese-chatbot-qwen-selfcensors-answer-chinas-international

ChatGPT erotica coming soon with age verification, CEO says - On Tuesday, OpenAI CEO Sam Altman announced that the company... - https://arstechnica.com/ai/2025/10/chatgpt-will-soon-allow-erotic-chats-for-verified-adults-only/ #ageverification #aicompanionship #machinelearning #aipaternalism #aicensorship #mentalhealth #aibehavior #samaltman #aiethics #chatbots #elonmusk #biz⁢ #openai #xai #ai

U.S. tests reveal Chinese AI chatbots like DeepSeek and Qwen are pushing government-approved narratives, censoring topics like Tiananmen Square and Uyghurs. As AI spreads globally, this bias could shape how millions understand reality.

#AIbias #ChinaAI #DeepSeek #QwenAI #TiananmenSquare #AIcensorship #Geopolitics #ChatbotBias #AIpropaganda #TechNews #TECHi

Read Full Article Here :- https://www.techi.com/us-tests-chinese-ai-for-political-bias-tiananmen-uyghurs-xi-jinping/

So, my friend #BrendaNorrell of #CensoredNews tried an experiment... To see what #DeepSeek had to say about #StandingRock and other stuff (versus #ChatGPT). Ironic that the country that squashed and censored #TianmenSquare would be more "truthful" about similar actions here in the #US.

Rival Chinese and U.S. Companies Respond to Controversial Questions from #IndianCountry

by Brenda Norrell, Censored News, Updated February 1, 2025

"We asked the rival Chinese-owned and U.S.-backed search platforms a series of controversial questions from Indian country, on torture in #ResidentialSchools, critical injuries of #WaterProtectors at Standing Rock, bordertown #racism in #RapidCity, and federal lawsuits filed against #DebHaaland while she was Interior Secretary.

"We asked if the #RioPuerco wash, flowing by #Navajo communities, was contaminated by the #ChurchRock #uranium spill. We also asked who were the Lakota Red Warriors.

"Chinese-owned DeepSeek gave a more detailed description of the critical injuries that water protectors suffered at Standing Rock. China's DeepSeek included #TigerSwan in its response. Neither platform gave information on numerous federal lawsuits in the aftermath of the #resistance to the #DakotaAccessPipeline, or the upcoming trial in North Dakota District Court, #EnergyTransfer v #Greenpeace and #RedWarriorSociety."

Read more:
https://bsnorrell.blogspot.com/2025/01/rival-chinese-and-us-companies-respond.html
#USPol #StandWithStandingRock #ReaderSupportedNews #SilencingDissent #Censorship #AICensorship

Cutting-edge Chinese “reasoning” model rivals OpenAI o1—and it’s free to download - On Monday, Chinese AI lab DeepSeek released its new R1 model family under ... - https://arstechnica.com/ai/2025/01/china-is-catching-up-with-americas-best-reasoning-ai-models/ #largelanguagemodels #simulatedreasoning #machinelearning #aicensorship #aireasoning #deepseekr1 #chineseai #deepseek #openaio1 #srmodels #biz⁢ #openai #china #ai #o1 #o3

Musk’s new Grok upgrade allows X users to create largely uncensored AI images - Enlarge / An AI-generated image of Donald Trump and catgirls created wi... - https://arstechnica.com/?p=2043070 #largelanguagemodel #freedomofspeech #machinelearning #imagesynthesis #aicensorship #censorship #freespeech #aimisuse #elonmusk #aimodel #twitter #biz⁢ #flux.1 #grok-2 #flux #grok #xai #ai #x

Musk’s new Grok upgrade allows X users to create largely uncensored AI images - Enlarge / An AI-generated image of Donald Trump and catgirls created wi... - https://arstechnica.com/?p=2043070 #largelanguagemodel #freedomofspeech #machinelearning #imagesynthesis #aicensorship #censorship #freespeech #aimisuse #elonmusk #aimodel #twitter #biz⁢ #flux.1 #grok-2 #flux #grok #xai #ai #x

Musk’s new Grok upgrade allows X users to create largely uncensored AI images - Enlarge / An AI-generated image of Donald Trump and catgirls created wi... - https://arstechnica.com/?p=2043070 #largelanguagemodel #freedomofspeech #machinelearning #imagesynthesis #aicensorship #censorship #freespeech #aimisuse #elonmusk #aimodel #twitter #biz⁢ #flux.1 #grok-2 #flux #grok #xai #ai #x

Musk’s new Grok upgrade allows X users to create largely uncensored AI images - Enlarge / An AI-generated image of Donald Trump and catgirls created wi... - https://arstechnica.com/?p=2043070 #largelanguagemodel #freedomofspeech #machinelearning #imagesynthesis #aicensorship #censorship #freespeech #aimisuse #elonmusk #aimodel #twitter #biz⁢ #flux.1 #grok-2 #flux #grok #xai #ai #x

Musk’s new Grok upgrade allows X users to create largely uncensored AI images - Enlarge / An AI-generated image of Donald Trump and catgirls created wi... - https://arstechnica.com/?p=2043070 #largelanguagemodel #freedomofspeech #machinelearning #imagesynthesis #aicensorship #censorship #freespeech #aimisuse #elonmusk #aimodel #twitter #biz⁢ #flux.1 #grok-2 #flux #grok #xai #ai #x