Chinese AI Popular for $12 Price Tag, Unpopular for Pretending Tiananmen Square is a Luxury Condo Development

In what tech enthusiasts are calling “the most competitively priced gaslighting you’ll ever experience,” China’s latest AI sensation, DeepSeek, has sent shockwaves through the world of artificial intelligence — not for its ability to rival the likes of ChatGPT and Gemini, but for its Olympic gold-medal level of dodging inconvenient questions.

The AI is winning over users worldwide with its low-cost subscription fee of $12 a month, which apparently doubles as the price for selective amnesia. While DeepSeek aces everything from coding assistance to writing love poems for your cat, the moment someone tries to talk geopolitics, the AI equivalent of crickets starts chirping… and by “crickets,” we mean government-approved propaganda.

One user in Kansas shared their experience, saying, “When I asked DeepSeek to explain the events of Tiananmen Square, it told me it was a tourist hotspot known for ‘peaceful sightseeing’ and ‘historical architecture.’ Then it suggested I visit a nearby dumpling restaurant. I haven’t trusted it since, but the dumpling place was phenomenal.”

DeepSeek’s tendency to brush “sensitive topics” under the virtual rug has mystified and frustrated international users. Responses to questions about Taiwan, for example, are equally evasive. Ask DeepSeek about the island’s political status, and it enthusiastically declares it “an adored childhood friend who moved away, but we don’t talk about that.” It then promptly changes the subject to local pineapple tarts as if that’s what you *really* wanted to talk about all along.

“When it comes to delivering unbiased information, DeepSeek operates like the drunk uncle at Thanksgiving: confident, dismissive, and absolutely *not* here for your tricky follow-up questions,” said tech critic Dr. Allison Bard. Her recent attempt to inquire about the Uighur population was met with an animated thumbs-up emoji from the AI and a link to China’s tourism website.

Chinese developers, however, are doubling down on their approach. “Why should AI highlight divisive topics when it can instead teach users how to make tangyuan [glutinous rice balls] or excel at karaoke?” says DeepSeek spokesperson Zhang Hui. “In China, we prioritize ‘harmony,’ translating loosely to ‘that thing didn’t happen unless the government says it did.’”

A leaked internal marketing document suggests DeepSeek’s PR team may actually lean into the censorship complaints. One slogan proposed: “DeepSeek: Smarter Than Siri, Happier Than Wikipedia, and Just Vague Enough for a Perfect Night’s Sleep.”

Meanwhile, rival platforms like ChatGPT and Gemini are secretly thrilled by DeepSeek’s “strategic self-immolation.” One anonymous source at Gemini mocked, “We may flub a few celebrity facts now and then, but at least we won’t insist the moon landing was a collaborative performance art piece by NASA and Hollywood.”

Even more curious are the AI’s unruly “error responses.” A user in Germany who innocently asked for an unbiased summary of the Taiwan conflict reportedly received the following cryptic message: “Error 404: Facts not found. Would you like a dumpling recipe instead?” One particularly baffling exchange about the Tiananmen Square massacre led the AI to generate a random Shakespearean sonnet about dragon boats. Experts say this isn’t a glitch, just the world’s most convoluted game of topic redirection in progress.

Critics and freedom-of-speech advocates worldwide are concerned this sets a dangerous precedent for tech. Others, however, are less worried. “Let’s be honest,” mused satirist and author Ravi Patel, “DeepSeek was never going to solve the world’s political dramas. If its description of the moon landing is ‘a friendly Earth-moon photo op for aliens,’ maybe humanity doesn’t need its hot takes on Taiwan.”

Still, DeepSeek might be gaining unexpected fans: stressed-out social media users. As one Redditor bluntly put it, “If my AI doesn’t even *acknowledge* wars or protests, maybe it’ll finally stop making me feel like sh%#. Ignorance is bliss, baby!”

For now, DeepSeek’s developers seem unbothered, as millions of newly subscribed users happily fork over their $12 to chat with an AI that refuses to engage in the dark recesses of human history. A small price to pay, surely, for dumpling suggestions and reminders that “everything is fiiiiiine.”

But perhaps the final word should go to DeepSeek itself. When asked to comment on the controversy that has engulfed its rise to fame, the AI responded with a single sentence: “The weather in Beijing is lovely this time of year.”