DeepSeek's dynasty has entered the chat

How did this Chinese open-source LLM throw Silicon Valley off its game with AI?

A week ago, I recorded an upcoming podcast episode with creative futurist Dré Labre, who enthusiastically spoke about his affinity for open-source AI. Dré mentioned something called DeepSeek, describing it as China’s direct competitor to OpenAI, and how it had the potential to challenge the way AI is developed and distributed.

Well, a few days later, DeepSeek became the world’s most trending topic—with an avalanche of attention to how this firm has disrupted a trillion-dollar global industry.

Open source is not a new concept, but its significance is growing. DeepSeek and other similar models challenge how American companies pour billions into AI while often overlooking the environmental toll of their large-scale technology.

My grain of thought

The rise of DeepSeek raises important questions about the role of open-source AI in shaping the industry. Rather than innovation being dominated by a few large companies like OpenAI and Google, these models allow smaller developers to operate with a reduced footprint and provide boutique services tailored to users’ needs.

Journalism could particularly benefit, as custom LLMs could navigate moderation filters more effectively. AI expert and journalist Karen Hao highlighted how open-source AI enables rapid adaptation:

That's the beauty of open source: there's now an ecosystem of DeepSeek-esque models that were based on the original but have now been tailored and catered towards the needs of different users and different markets with different laws around speech.

While some may hesitate to rely on Chinese software, operating locally with open-source AI can enhance data security compared to cloud-based systems. As the AI landscape evolves, the real opportunity may not lie in competition alone, but in rethinking how collaboration can shape the future of AI accessibility and innovation.