Deepseek and its sudden rise, overthrowing ChatGPT, how?

A person’s silhouette faces a screen displaying DeepSeek’s blue whale logo. A laptop in the foreground shows the same logo with the text "deepseek."

DeepSseek has experienced an overwhelming surge in sign-ups, surpassing the initial traction seen by ChatGPT (OpenAI) when it was first introduced. However, this phenomenon is not solely attributable to DeepSeek’s coding, quantitative reasoning, general reasoning, or knowledge acquisition capabilities. The answer is not a straightforward “yes” or “no.”

Rather, this is a classic case of a causal effect in action. The widespread adoption of GPT-based models had already familiarized users with large language models (LLMs) before DeepSeek entered the market. As these technologies became increasingly accessible and affordable, curiosity naturally drove users to explore and experiment with the latest offerings. The adoption rate might have been even higher if not for concerns regarding its Chinese origin, which raises questions about potential government influence.

Ultimately, this surge in sign-ups results from the momentum built by earlier LLMs, making it an expected development rather than an anomaly. The real question now is: what comes next?

DeepSeek’s rapid adoption can be understood through the lens of the Technology Adoption Lifecycle, which consists of five customer segments: Innovators, Early Adopters, Early Majority, Late Majority, and Laggards. Understanding how these segments interact and market saturation is perceived is essential for understanding how a new AI model like DeepSeek can achieve full adoption. This analysis of market dynamics is crucial.


Diffusion of innovations graph showing adoption stages: Innovators (2.5%), Early Adopters (13.5%), Early Majority (34%), Late Majority (34%), and Laggards (16%).
  1. Innovators (2.5%) – The Pioneers
    This segment comprises tech enthusiasts and AI researchers actively seeking the latest advancements, regardless of potential flaws. These individuals were already familiar with large language models (LLMs) and eagerly tested DeepSeek’s capabilities, particularly its coding, reasoning, and affordability strengths.
  2. Early Adopters (13.5%) – The Trendsetters
    This group includes AI, software development, and data science professionals who rely on cutting-edge tools to enhance productivity, gain competitive advantages, and achieve cost efficiency. Their adoption of DeepSeek influences subsequent segments, as they help spread awareness and credibility.
  3. Early Majority (34%) – The Pragmatists
    As DeepSeek gains credibility through positive word-of-mouth, the Early Majority—comprising businesses, educators, and casual users—will begin to adopt it. This segment requires proven value and stability before transitioning from established platforms like ChatGPT. Their adoption will be a significant growth driver, especially if DeepSeek is perceived as offering superior features or affordability.
  4. Late Majority (34%) – The Skeptics
    This group is more resistant to change and typically waits until a technology becomes the industry standard. If DeepSeek establishes itself as a mainstream AI tool, corporate environments and educational institutions will adopt it out of necessity rather than innovation. Strategic partnerships, enterprise integrations, and regulatory acceptance will be essential for this group.
  5. Laggards (16%) – The Holdouts
    Laggards are resistant to change and prefer traditional methods or older technologies. However, as AI-powered tools become increasingly unavoidable in both professional and personal contexts, even this group will eventually adopt DeepSeek, often due to external pressures such as workplace requirements or the diminishing availability of alternatives.

It basically means that the first 4 from Innovators, Early Adopters, Early Majority and even the Late Majority, and even high proportions of Laggards are all collapsing into a one-time dimension that catapulted Deepseek to the top, just like when OpenAI came off with ChatGPT, only lagging behind now.

Summary

DeepSeek’s unprecedented surge in sign-ups has outpaced the initial adoption of ChatGPT by OpenAI. However, this phenomenon is not solely due to DeepSeek’s coding, quantitative reasoning, or general knowledge acquisition capabilities. Instead, it reflects a broader causal effect driven by the prior mainstream adoption of GPT-based models. With large language models (LLMs) already integrated into everyday workflows, introducing a new, more affordable alternative naturally attracted users eager to explore its potential. Concerns regarding its Chinese origin and potential government oversight may have tempered an even greater adoption rate.

The rapid adoption of DeepSeek aligns with the Technology Adoption Lifecycle, where five distinct customer segments—Innovators, Early Adopters, Early Majority, Late Majority, and Laggards—typically drive market expansion. Innovators and early adopters, including AI researchers and technology professionals, were the first to test DeepSeek’s capabilities. Their influence spurred adoption among the early majority, including businesses and educators, who require proven functionality before transitioning from established platforms like ChatGPT.

The late majority, often hesitant to adopt new technology, will likely embrace DeepSeek as it becomes a market standard. In contrast, laggards—typically the most resistant—will adopt it only when external pressures make it unavoidable.

In this case, however, the traditional staged adoption process appears to have collapsed into a single accelerated phase, similar to the early rise of ChatGPT. The convergence of these adoption segments has propelled DeepSeek to prominence in a compressed timeframe, bypassing the gradual progression seen in conventional technology adoption models. This shift suggests that LLM adoption is reaching a level of ubiquity where market leaders can rapidly change, making the long-term dominance of any one model far from guaranteed. The key question is what the next breakthrough will be and how the AI landscape will continue evolving.
Written by Dr. Mak Wai Keong

No responses yet

Leave a Reply

Your email address will not be published. Required fields are marked *