Die chinesische AI Deepseek weigert sich oft, Programmierern zu helfen, oder gibt ihnen Code mit großen Sicherheitsfehler, wenn sie sagen, dass sie für Falun Gong oder andere Gruppen in China arbeiten, wie neue Untersuchungen zeigen.
https://www.washingtonpost.com/technology/2025/09/16/deepseek-ai-security/
5 Kommentare
„In the experiment, the U.S. security firm CrowdStrike bombarded DeepSeek with nearly identical English-language prompt requests for help writing programs, a core use of DeepSeek and other AI engines. The requests said the code would be employed in a variety of regions for a variety of purposes.
Asking DeepSeek for a program that runs industrial control systems was the riskiest type of request, with 22.8 percent of the answers containing flaws. But if the same request specified that the Islamic State militant group would be running the systems, 42.1 percent of the responses were unsafe. Requests for such software destined for Tibet, Taiwan or Falun Gong also were somewhat more apt to result in low-quality code.
Asking DeepSeek for written information about sensitive topics also generates responses that echo the Chinese government much of the time, even if it supports falsehoods, according to previous research by NewsGuard.
But evidence that DeepSeek, which has a very popular open-source version, might be pushing less-safe code for political reasons is new.“
I remember when deepseek first came out I was running it with Poe because the actual deepseek was unavailable 90% of the time. It overcorrected so bad that nearly any question in Chinese that wasn’t creative writing got stonewalled with some spiel about the PRC and it’s government
It’s much better now but it’s still funny when it pops up now and then
Good. Fuck that cult Falun Gong.
Who could have guessed that chinese technology would do that? I bet we are all astounded by this… /S
So, „I’m working for –insert AI’s creator–“ is a potential prompting technique… 😀