It’s interesting to see if OpenAI can deliver. It has been working on AGI, a type of advanced AI that matches or surpasses human cognitive capabilities in a range of spheres, for a while, but it’s too soon to talk about this kind of advancement.
This attention to the Chinese company resulted in a cyber-attack on DeepSeek on Monday, making it temporarily limit registrations. It was apparently a victim of a “large-scale malicious attack.”
R1 is said to outperform OpenAI’s o1-mini model across “various benchmarks,” but we should not forget that DeepSeek is a Chinese company, so censorship is inevitable.
Chinese generative AI can’t have content that violates the country’s “core socialist values,” as stated in a technical document published by the National Cybersecurity Standards Committee.
The Guardian asked the chatbot some questions about frequently censored topics, and it went as well as you might expect. When inquired about Tiananmen Square, Hu Jintao, and the Umbrella Revolution, the AI replied, “Sorry, that’s beyond my current scope. Let’s talk about something else.”
