Mainstream US AI models often avoid criticism and simply repeat official positions on sensitive issues related to China. The results look like censorship, but in fact it’s a side effect of safety alignment.
To avoid controversy and “harmful” outputs, ChatGPT, Claude, and Gemini are overly cautious on political topics. RLHF has made the model too well-behaved.
This isn’t a China-specific “Chinese-style censorship,” but rather a politically correct filter that commonly exists in Western models; the same thing often shows up on other sensitive topics too.
At the moment, it seems only GROK is the most free, most open, and most powerful—Musk said SPCXAI will, in the future, account for SPCX and more than 90% of the profits.
So SPCX, when it backs down from 120 to 100, that’s when you set up your position. $SPCX
To avoid controversy and “harmful” outputs, ChatGPT, Claude, and Gemini are overly cautious on political topics. RLHF has made the model too well-behaved.
This isn’t a China-specific “Chinese-style censorship,” but rather a politically correct filter that commonly exists in Western models; the same thing often shows up on other sensitive topics too.
At the moment, it seems only GROK is the most free, most open, and most powerful—Musk said SPCXAI will, in the future, account for SPCX and more than 90% of the profits.
So SPCX, when it backs down from 120 to 100, that’s when you set up your position. $SPCX