DeepSeek R1 can generate more harmful content than other AI, study finds
A security report shows that DeepSeek R1 can generate more harmful content than other AI models without any jailbreaks.
What questions will China's DeepSeek not answer?
The DeepSeek AI chatbot becomes tongue-tied when asked about issues seen as politically sensitive by China's Communist Party. DW tried it out.
Daily Wrap on MSN40m
Beijing boosts DeepSeek's AI launch, challenges U.S. tech
According to the analysis firm Graphika, Chinese social media accounts linked to the government in Beijing supported the launch of DeepSeek's chatbot, as reported by Reuters on Friday. Among these ...
adarshbadri42m
Things China’s DeepSeek Does Not—And Will Not—Tell You About Politics
Founded two years ago, in 2023, the Chinese AI company DeepSeek had already achieved what would have seemed almost impossible.