Flavio Coelho/Getty Images
Just a few weeks after gaining widespread recognition, Chinese AI startup DeepSeek is rapidly moving ahead of its competitors and sparking discussions about the benefits of open-source software. However, concerns about security have emerged surrounding the company, leading private and government organizations to prohibit the use of DeepSeek. Here’s what you should know.
What is DeepSeek?
Established by Liang Wenfeng in May 2023, DeepSeek is a Chinese startup that has quickly challenged established AI companies with its open-source approach. The company’s funding model, supported by the High-Flyer hedge fund run by Wenfeng, has enabled fast growth and research initiatives. DeepSeek gained attention when it unveiled R1, its open-source reasoning model that outperformed OpenAI’s o1. The AI assistant powered by DeepSeek’s V3 model surpassed the popularity of ChatGPT, becoming one of the most downloaded free apps on the App Store. DeepSeek has been praised for its ability to create successful models by optimizing older chips strategically and distributing query loads efficiently.
What is DeepSeek R1?
Released in January, R1 is DeepSeek’s flagship reasoning model that excels in math, coding, and reasoning benchmarks. Unlike models from tech giants, R1 is open source, allowing anyone to download and use it. DeepSeek’s pricing is more affordable compared to similar US models, making it an attractive option for many users. However, concerns have been raised about censorship in Chinese models, which could impact the long-term competitiveness of DeepSeek against US alternatives.
Privacy and Security Concerns
Data privacy worries similar to those associated with TikTok have emerged regarding DeepSeek. Security experts have discovered links to Chinese government-controlled servers, raising concerns about user data being sent to China. Flaws in DeepSeek’s mobile app, including unencrypted data and poor data storage practices, have also been identified. These security issues highlight the importance of safeguarding user data when using AI platforms.
Safety Concerns
AI safety researchers have expressed concerns about the potential risks associated with powerful open-source models like DeepSeek R1. While US AI companies typically have safety oversight teams, it is unclear if DeepSeek has a similar setup in place. The lack of transparency regarding safety measures raises concerns about the potential misuse of AI technology.
In conclusion, while DeepSeek’s innovative approach to AI has garnered attention in the industry, ongoing security and safety concerns highlight the need for robust safeguards to protect user data and ensure responsible use of AI technology.