Gartner Forecasts Increased Use of Task-Specific AI Models by 2027

According to Gartner, by 2027, organisations are expected to adopt small, task-specific AI models at a scale three times greater than general-purpose large language models (LLMs).

While general-purpose LLMs offer strong language capabilities, their accuracy tends to decrease when applied to tasks that require a detailed understanding of specific business domains. As a result, enterprises are likely to prioritise specialised models that deliver more precise and context-aware outputs for targeted use cases.

“The variety of tasks in business workflows and the need for greater accuracy are driving the shift towards specialised models fine-tuned on specific functions or domain data,” said Sumit Agarwal, VP Analyst at Gartner. “These smaller, task-specific models provide quicker responses and use less computational power, reducing operational and maintenance costs.”

Gartner Study - Enterprises Leverage RAG and Fine-Tuning to Develop Task-Specific LLMs

To create specialised language models for specific tasks, enterprises can apply retrieval-augmented generation (RAG) or fine-tuning techniques. These methods enable the customisation of large language models using internal data sources.

In this process, enterprise data serves as a critical differentiator, requiring structured preparation. This includes conducting quality checks, implementing version control, and managing data to ensure it meets the requirements for fine-tuning and supports the accuracy and relevance of the resulting models.

“As enterprises increasingly recognise the value of their private data and insights derived from their specialised processes, they are likely to begin monetising their models and offering access to these resources to a broader audience, including their customers and even competitors,” said Agarwal. “This marks a shift from a protective approach to a more open and collaborative use of data and knowledge.” 

Commercialising Proprietary Models and Deploying Task-Specific AI

Enterprises can generate new revenue opportunities by commercialising their proprietary language models while also contributing to a more connected and collaborative AI ecosystem.

Recommendations for Implementing Task-Specific AI Models

Organisations aiming to deploy small, task-specific AI models should consider the following steps:

• Pilot Contextualised Models: Deploy smaller, domain-specific models in business areas where contextual understanding is critical or where general-purpose large language models have not met quality or latency requirements.

• Adopt Composite Model Approaches: Identify scenarios where a single model is insufficient and adopt composite approaches that coordinate multiple models and workflow stages to achieve better outcomes.

• Invest in Data and Skills Development: Focus on preparing enterprise data by collecting, curating, and structuring it for fine-tuning. At the same time, upskill teams across key functions—including AI and data architecture, data science, engineering, risk, compliance, procurement, and subject matter experts—to ensure the successful execution and governance of these AI initiatives.
  
  

Read More:

We Emphasise on Interdependence of AI and Cloud technologies

Servers and Data Storage both are the Pillars

CRM Innovations in the Mobile Segment

Integrating AI, data analytics, and Technologies Help Businesses