The Rise of LLMs and the Current State of the AI Market

In the early 2000s, artificial intelligence was largely dominated by rule-based systems and supervised learning, resulting in domain-specific or narrow AI solutions. However, the introduction of the Transformer architecture by Google in 2018, followed by the success of OpenAI’s GPT series (GPT-2 and GPT-3), marked the beginning of a new era of general-purpose AI driven by large language models. These models, trained on massive datasets and built with billions of parameters, have demonstrated near human-level performance in tasks such as natural language generation, translation, summarization, and coding. This has significantly broadened the scope of AI applications.

Today, services based on large language models are rapidly becoming mainstream, attracting hundreds of millions of users around the world. OpenAI’s ChatGPT surpassed 100 million monthly active users within just two years of its launch. Microsoft Copilot has been integrated across its suite of office software, impacting the workflows of tens of millions of users. Additionally, models such as Google Gemini, Claude, and open-source models based on Meta’s LLaMA are being adopted across various industries. AI is now being used in everyday settings, including customer support, search, education, finance, software development, law, and healthcare.

The widespread adoption of LLMs has not only accelerated the growth of the AI market but also driven demand for more specialized and segmented AI solutions. Emerging trends include customized AI services, on-device AI powered by edge computing, and privacy-preserving AI. These developments are creating new architectures and market opportunities that complement and extend beyond centralized large-scale models.