Member-only story
OpenAI, DeepSeek, and Qwen 2.5
The 2025 AI Race and the Struggle for Efficiency
How the AI disruption story unfolds

In this blog post, I briefly touch the surface of he dynamic journey of AI innovation — examining OpenAI’s grand investments, DeepSeek’s disruptive cost-efficiency, and Qwen 2.5’s intelligent optimization — to reveal how each approach is reshaping the future of AI.
The start
Since its inception in 2015, OpenAI has established itself as a leading force in the development of advanced AI models. With billions invested and collaborations with major tech companies such as Microsoft and SoftBank, the organization has built powerful models like GPT-4.
Big Investments
Recently, the United States announced a $500 billion investment in the Stargate project — a partnership involving OpenAI, Microsoft, Oracle, and MGX. This initiative primarily focuses on expanding GPU-equipped datacentres that are going to empower the U.S. AI capacity. OpenAI CEO Sam Altman has remarked that it is practically impossible to catch up with OpenAI given the enormous costs and infrastructure required to build competitive models.