DeepSeek has taken the tech world by storm for the past month after it was revealed that the company’s models were trained for a fraction of the cost of other top models while still delivering competitive performance in many areas. Not only were they cheaper to train, but they’re also cheaper to run, making DeepSeek’s models an attractive option for developers looking to reduce AI expenses.
DeepSeek was created in China, and some companies and government have expressed concerns about where their data is being stored and how it’s being used, with some even banning its use. However, the models are open source, so anyone can download them and run them offline locally on their own devices, which can mitigate some of the security and privacy concerns.
To explore the implications of DeepSeek for developers, Jenna Barron, news editor of SD Times, spoke to a panel of industry. Watch now to hear insights from Melissa Ruzzi, director of AI at AppOmni; Bratin Saha, chief product and technology officer at DigitalOcean; and Kate O’Neill, author, speaker, and executive consultant on technology’s impact on the human experience.
