Innodata Inc. has signed a multi-year agreement with Palantir Technologies to deliver specialized data engineering and high-quality training data. Under the deal, Innodata’s annotation teams will process over 5,000 hours of high-resolution rodeo video footage to train computer vision algorithms capable of identifying animals, riders and detailed skeletal movements. The scope includes multimodal data integration, quality assurance workflows and generative-AI support to optimize labeling efficiency and accuracy.

Following the announcement, Innodata’s shares surged 14.4% on heavy trading, nearly doubling its three-month average daily volume. Investors responded to the validation of Innodata’s AI content services by a leading enterprise software provider. Since the contract reveal, institutional buyers have increased their holdings, driving a pick-up in liquidity and narrowing the bid-ask spread.

The partnership is expected to contribute approximately $8 million in incremental revenue over the next 12 months, representing nearly 10% of Innodata’s current annual revenue run rate. Management forecasts that the high-value annotation services will carry gross margins in excess of 35%, bolstering overall profitability. Incremental operating leverage from scalable platform enhancements could lift adjusted EBITDA margins by 200 basis points in fiscal year 2026.

Analysts highlight the deal as a key proof point for Innodata’s repositioning toward AI-enabled content solutions. Several brokerage firms have revised their 2026 revenue forecasts upward by 5%–8%, citing repeatable deployment potential across Palantir’s government and commercial clients. While short-term estimate revisions may moderate, the strategic alliance is viewed as a durable growth driver that validates Innodata’s investments in automation and machine learning infrastructure.