Global end-user spending on generative AI is projected to hit $14.2 billion in 2025, driven by rising demand for domain-specific models and open-source alternatives. According to Gartner, over 50% of enterprise GenAI models will be domain-specific by 2027, with sustainability and autonomous AI also shaping the future of the technology.
Global spending by end users on generative AI (GenAI) technologies is projected to reach a staggering $14.2 billion in 2025, according to a new report from research and advisory firm Gartner. This figure reflects the continued surge in investment in GenAI tools, platforms, and services across industries as organizations accelerate digital transformation efforts and adopt intelligent automation.
Of this total, Gartner estimates that $1.1 billion will be spent specifically on specialized generative AI models — known as domain-specific language models (DSLMs). These models are trained or fine-tuned on highly specific data related to particular industries or workflows, making them more efficient and relevant for targeted use cases compared to more general foundation models.
“Foundation GenAI models, including large language models (LLMs), are trained on vast datasets and can perform a wide range of tasks,” said Arunasree Cheparthi, Senior Principal Research Analyst at Gartner. “They will continue to dominate GenAI spending due to their broad applicability. However, we are increasingly seeing enterprises turn to domain-specific models that offer better performance, reduced cost, and higher relevance in focused applications.”
Table of contents [Show]
Gartner predicts a dramatic rise in the use of domain-specific GenAI over the next few years. By 2027, more than 50% of GenAI models in enterprise use will be domain-specific, up sharply from just 1% in 2024. These DSLMs, while typically smaller in size compared to foundation models like GPT-4, are proving to be more cost-effective and impactful in industry-specific scenarios such as finance, healthcare, legal, and manufacturing.
This shift underscores a broader trend in GenAI development: moving from general-purpose models to more specialized, optimized solutions that meet unique business needs. As a result, many organizations are investing in models that align closely with their data environments, regulatory constraints, and customer expectations.
As the GenAI ecosystem matures, sustainability is becoming a key consideration. Gartner forecasts that by 2028, 30% of GenAI deployments will be optimized for energy efficiency, driven largely by corporate sustainability goals and mounting pressure to reduce AI’s environmental footprint. This will likely lead to innovations in hardware design, data center optimization, and algorithmic efficiency.
Furthermore, Gartner projects that by 2028, one-third of GenAI interactions will involve autonomous agents or action models capable of completing complex tasks from start to finish with minimal human intervention. These advanced systems mark a step closer to artificial general intelligence (AGI) — a concept still theoretical, but increasingly the subject of serious academic and commercial research.
Another emerging trend is the growing popularity of open-source GenAI models, which are increasingly viewed as viable alternatives to proprietary systems. As regulatory scrutiny around AI intensifies globally, organizations are exploring open-source solutions for their flexibility, transparency, and potential for customization.
“Open-source GenAI is becoming more attractive, especially in highly regulated sectors,” said Cheparthi. “Companies value the ability to audit, modify, and control their AI systems more tightly, especially when dealing with sensitive data and compliance obligations.”
Despite the explosive interest in GenAI, experts caution businesses against overestimating its short-term impact. Many GenAI solutions have already reached the "Peak of Inflated Expectations" in Gartner’s Hype Cycle™, and successful implementation often requires careful planning, high-quality data, and skilled talent.
Nonetheless, adoption is accelerating. By 2026, Gartner predicts that 75% of businesses will use GenAI to generate synthetic customer data — a massive leap from less than 5% in 2023.
As generative AI continues to evolve, its influence will be felt across nearly every sector of the global economy. From smarter chatbots and virtual assistants to domain-optimized tools and autonomous systems, the future of enterprise AI is shaping up to be both intelligent and industry-aware.
