DeepSeek, a Chinese artificial intelligence startup, has been praised by Google DeepMind CEO Demis Hassabis as 'probably the best work' he has seen from China in the AI field. However, Hassabis also stated that the hype around DeepSeek's achievements is 'exaggerated' and that the company's AI model, known as R1, does not represent a new scientific advance, but rather uses known techniques. DeepSeek's R1 model has been noted for its efficiency and was developed at a reported cost of under $6 million, a claim Hassabis described as 'exaggerated and a little bit misleading.' The startup's success has sparked a significant reaction in the global AI community, with major tech companies and investors reevaluating their strategies in light of DeepSeek's advancements. DeepSeek is also aggressively recruiting specialists in the field of Artificial General Intelligence (AGI), indicating its ambition to push the boundaries of AI technology further.