EXL Boosts AI
EXL acquires AI training leader iMerit for up to $310 million to enhance enterprise solutions.
Key Points
EXL's acquisition of iMerit includes upfront cash of $170 million with the potential for up to $140 million in earnouts, and a portion of the consideration will be held in escrow for working capital and indemnifiable matters; the closing is targeted for Q3 2026 and subject to customary closing conditions and antitrust clearance, with termination rights available to both sides.
iMerit brings direct relationships with foundation model builders, providing EXL with early visibility into how foundation models are trained, fine-tuned and improved for enterprise use.
The deal designates Clairvoyant AI, Inc., an indirect wholly owned subsidiary of EXL, as the purchaser of iMerit, highlighting the transaction's internal structuring within EXL.
EXL's leadership frames AI development with a focus on reliable enterprise outcomes, with Rohit Kapoor stating that success requires industry-specific data, rigorous evaluation and reinforcement learning to deliver reliable results in business-critical workflows.
EXL Service Holdings has agreed to acquire iMerit, an AI model training company, for up to $310 million, the company announced June 24, 2026. The deal includes $170 million in upfront cash and up to $140 million in additional earnouts tied to performance over two years, according to Nasdaq.
The purchase will be made through Clairvoyant AI, Inc., an EXL subsidiary. The transaction is expected to close in Q3 2026, pending antitrust clearance and other standard conditions, according to Let's Data Science.
iMerit Gives EXL a Seat at the AI Training Table
iMerit, founded in 2012, specializes in training and evaluating AI models. It works directly with foundation model builders — companies like OpenAI and Anthropic that build large AI systems. That gives EXL early visibility into how these models are built and improved, according to GuruFocus.
iMerit also brings two key tools: its Ango data platform and its Scholars network, a group of specialized human annotators including doctors and engineers. These experts provide the human feedback that AI models need to improve — a process called reinforcement learning from human feedback, or RLHF.
EXL's CEO Says AI Needs Industry-Specific Data to Work
EXL Chairman and CEO Rohit Kapoor framed the deal around a core problem: generic AI models often fail in real business settings. "Success requires industry-specific data, rigorous evaluation and reinforcement learning to deliver reliable results in business-critical workflows," Kapoor said, according to Wahanariau.
iMerit CEO Radha Ramaswami Basu echoed that view. "We see EXL as an ideal leader in this defining moment for AI," she said. "We can build on our work with AI innovators and bring those insights to companies seeking to unlock their proprietary data." The deal targets regulated industries like healthcare, finance, and autonomous vehicles.
$310M Price Tag Carries a $140M Performance Gamble
Nearly half the deal's value is not guaranteed. The $140 million in earnouts — 45% of the total $310 million — depends on iMerit hitting specific integration and performance targets over two years, according to Let's Data Science. Major iMerit backers, including Khosla Ventures, Omidyar Network, and British International Investment, will exit through the transaction.
EXL's stock has fallen nearly 40% in the first half of 2026, reaching a 52-week low of $25.18. Shares traded around $26.16 at the time of the announcement. Despite the decline, seven analysts recently revised earnings estimates upward, and EXL's price-to-earnings ratio of 16.68 sits near a 10-year low, according to Finviz.
Deal Reflects Broader Rush to Own AI's Building Blocks
The acquisition follows EXL's 2024 purchase of ITI Data. That deal focused on managing complex datasets. The iMerit deal goes further — moving EXL from data management into model training itself. A March 2025 McKinsey report found that 78% of organizations now use AI in at least one business function, up from 72% in 2024, creating growing demand for high-quality training data.
EXL says the deal will expand its total addressable market into high-growth AI sectors. One key focus: building smaller, cheaper AI models tailored for specific industries, known as small language models, or SLMs. These are increasingly popular with enterprises that want AI that is more secure and less expensive to run than large general-purpose models, according to GuruFocus.
Sentiment
Mostly PositivePublishers
10
Articles
34
Reach
44
