In a remarkable shift in the AI landscape, Chinese mobile application DeepSeek has surpassed OpenAI's ChatGPT to claim the top spot on the iPhone free app charts in both China and the United States. This significant achievement comes shortly after its developer unveiled the innovative \"reasoning model,\" DeepSeek R1.
DeepSeek R1 introduces a \"DeepThink\" mode, allowing users to witness the app's \"thinking process\" as it generates responses. This feature enhances the app's ability to tackle complex logical and mathematical problems, providing a more transparent and interactive user experience.
According to DeepSeek's official website, the R1 model's performance is \"on par with\" OpenAI-o1, yet operates at approximately one-thirtieth of the cost. The app and its web-based chatbot are currently free for general users, with only programmers needing to pay for API access. Moreover, DeepSeek has made its full-size model available for free download, enabling users with adequate hardware to run it locally. For those with less powerful devices, scaled-down versions optimized for various systems are offered.
In addition to providing free access to its models, DeepSeek has published a research paper detailing the development of R1. This transparency allows other developers to replicate the process using their own training data, fostering an environment of collaboration and innovation within the AI community.
Industry Leaders React
The release of DeepSeek R1 has garnered significant attention from prominent figures in the AI industry.
Marc Andreessen, venture capitalist and co-founder of Netscape, described DeepSeek R1 as \"one of the most amazing and impressive breakthroughs\" he has ever seen, calling it \"a profound gift to the world.\"
Aravind Srinivas, CEO of Perplexity, noted that \"DeepSeek has largely replicated OpenAI-o1-mini and has open-sourced it,\" highlighting the significance of the model's accessibility.
Jim Fan, senior research manager at Nvidia, praised DeepSeek as a company that is upholding the mission of conducting \"truly open, frontier research that empowers all.\"
Yann LeCun, Meta's chief AI scientist, emphasized that DeepSeek's success demonstrates how \"open-source models are surpassing proprietary ones,\" rather than signifying any national competition in AI leadership.
Kai-Fu Lee, former president of Google China, expressed that the success of DeepSeek validates his belief in China's potential to excel in generative AI engineering.
DeepSeek CEO on Innovation in China
In a July 2024 interview with 36kr.com, DeepSeek CEO Liang Wenfeng discussed the importance of China moving beyond the stereotype of merely applying foreign innovations. He emphasized that as China's economy grows, the country should become a global contributor to innovation.
\"Innovation stems not only from the pursuit of business success but also from genuine curiosity,\" Liang stated. He highlighted that the DeepSeek team consists of young talent, and the development of the earlier V2 model did not involve overseas Chinese contributors. \"Perhaps the top 50 talents in this field are not in China,\" Liang acknowledged, \"but we can cultivate our own.\"
DeepSeek's ascent to the top of the app charts signifies a notable shift in the AI industry, showcasing the rise of open-source models and the growing influence of Chinese innovators. As DeepSeek continues to develop and share its advancements, the global AI community watches with keen interest.
Reference(s):
cgtn.com
