How DeepSeek and Next-generation aI Agents might Erode Value Of Language Models

In the internet revolution, we’re moving from constructing sites as the main service to really developing – so, the Airbnb of AI, the Stripe of AI,” he added. “They are not about the model. They are about the system and how you make the design useful for tasks.”

The arrival of DeepSeek’s R1 model last week is seen by some tech CEOs as a more sign LLMs are becoming significantly commoditized.

R1 integrates some essential elements that separate it from other models on the market. For instance, the design incorporates a “blended precision” structure that uses a mix of full-precision 32-bit floating point (FP32) numbers and low-precision 8-bit floating point (FP8) numbers.

The latter is faster to process but can be less accurate. However, instead of rely on one or the other, DeepSeek uses FP8 for a lot of estimations and changes to FP32 for certain tasks where a higher degree of accuracy is required.

“As AI gets more effective and accessible, we will see its usage skyrocket, turning it into a commodity we simply can’t get enough of,” Microsoft CEO Satya Nadella wrote on social media platform X, in an obvious reference to DeepSeek.

Meanwhile, Matt Calkins, CEO of U.S. software company Appian, told CNBC that DeepSeek’s success simply reveals that AI models are going to become more of a product in the future.

“In my viewpoint, we’re going to see a commoditization of AI. Many business will achieve competitive AI, and an absence of differentiation will be bad for high-spending first-movers,” Calkins stated by means of e-mail.