A new A.I. model, released by a scrappy Chinese upstart, has rocked Silicon Valley and upended several fundamental assumptions about A.I. progress.The latest A.I. model from a Chinese start-up, DeepSeek, has caused a stir in Silicon Valley and challenged long-held beliefs about A.I. progress. The paper, titled “Incentivizing Reasoning Capability in LLMs via Reinforcement Learning,” was released last week and initially went unnoticed. However, as researchers delved into its claims, they were shocked by the implications. DeepSeek had created a new A.I. model, DeepSeek-R1, using a small number of lower-quality chips, yet it matched the performance of top American models at a fraction of the cost. This was achieved through clever engineering, challenging the notion that raw computing power is necessary for A.I. advancement. The fact that this breakthrough came from China, a country often seen as lagging behind in the A.I. race, added to the disbelief. Some speculated that DeepSeek had cheated or manipulated their results, but as more people examined the open-source software, their skepticism turned to concern. When the DeepSeek mobile app became the top app on Apple’s App Store, panic set in. The New York Times reports that the A.I. breakthrough has caused a frenzy in the tech world, with many questioning America’s dominance in the field.
Source:Read More