Generative AI adoption is the process by which organizations experiment with, and pilot, GenAI initiatives. Here are highlights from our recent survey.
Assessing generative AI adoption in enterprises
Generative AI (GenAI) is transforming the enterprise landscape and creating new possibilities for innovation, efficiency, and customer engagement. As organizations race to seize the potential of this game-changing technology, they also face significant challenges to generative AI adoption.
A recent Gartner report estimates that 30% of all generative AI projects will be abandoned in 2025 due to issues with AI data quality, inadequate risk controls, escalating costs, and unclear business value. This prediction underscores the critical role GenAI data plays in successful generative AI adoption.
To gain a clearer understanding of the key challenges organizations face today, we surveyed 300 enterprise professionals who play a direct role in planning, building, or delivering GenAI solutions in their organizations. This article includes select findings from the survey and also discusses Retrieval-Augmented Generation (RAG), a GenAI framework that’s becoming increasingly important for the enterprise.
Get your free copy of our State of GenAI Data Readiness report here.
From experimentation to pilots in GenAI adoption
Organizations are at various stages of their generative AI adoption journeys. Our survey revealed that 43% of respondents are engaged in GenAI pilot projects at the department or business unit level, while 13% are conducting company-wide pilots.
Only 2% have progressed to full production deployment of one or more generative AI use cases. Additionally, 17% are experimenting with GenAI at the business unit level, while 21% are doing so across their entire organization.
These findings suggest that while widespread production use is still limited, a significant number of enterprises are doing more than just experimenting. This trend aligns with the broader industry movement, where businesses are cautiously but proactively testing GenAI feasibility and value through pilot projects before committing to larger-scale deployments.
Current approaches to GenAI adoption
The RAG revolution
While an enterprise LLM (Large Language Model) offers impressive capabilities, it often falls short of meeting specific business needs without customization. Our survey shows that only 14% of organizations utilize off-the-shelf LLMs in their AI projects, while 86% are enhancing their LLMs by evaluating RAG vs fine-tuning vs prompt engineering in different combinations.
Leading LLM approaches in use today
By augmenting LLMs with relevant internal data, companies can significantly improve the performance and accuracy of their AI models, especially for customer-facing and operational applications. This approach allows organizations to tailor AI solutions to their unique contexts, ensuring that the technology delivers meaningful and actionable insights.
RAG architecture is rapidly gaining traction as a method for enhancing generative AI models with internal company data. Our survey found that 60% of respondents are currently piloting RAG, 24% are in the planning stages, and 12% have moved to full production implementation. A small portion (4%) remains in the exploratory phase.
Industry-specific adoption rates provide further insights:
RAG in piloting stage, by industry
These figures demonstrate strong momentum in adopting RAG for structured data, particularly in sectors where data privacy compliance and precise responses are critical. However, the limited number of enterprises that have transitioned to full production reflects the complexities involved in scaling these advanced technologies effectively.
Enterprises are ripe for generative AI adoption
Generative AI adoption is accelerating as enterprises recognize its transformative potential. Companies are strategically moving from experimentation to pilot projects, focusing on areas like marketing, sales, and customer operations where generative AI can deliver substantial benefits. Techniques like RAG are becoming essential tools for customizing AI models to meet specific business requirements.
While challenges remain – particularly regarding data quality, risk management, and scalability – the forward momentum is clear. Companies that navigate these hurdles effectively stand to gain a significant competitive advantage.
Discover GenAI Data Fusion RAG tools by K2view,
the enterprise choice for generative AI adoption.
