Tempus AI Launches Paige Predict to Predict 123 Biomarkers from Single H&E Slides

TEMTEM

Tempus AI launched Paige Predict, a pathology suite analyzing H&E slides to predict 123 biomarkers across 16 cancer types using data from 200,000 de-identified patient samples. It informs clinicians on test sequencing to avoid tissue-scarcity delays by providing actionable molecular insights from a single slide image.

1. Ark Invest Increases Stake in Tempus AI

On January 22, 2026, Cathie Wood’s Ark Invest purchased a combined total of 13,532 shares of Tempus AI across its Genomic Revolution ETF and Innovation ETF, representing approximately $2.4 million of new investment in the company. These purchases follow a larger acquisition the previous day, when Ark Invest deployed $5.85 million to add to its Tempus AI holdings. The firm’s renewed conviction comes after Tempus AI reported an 83% year-over-year increase in total revenue for 2025, driven by a 111% surge in diagnostics revenue, signaling accelerating commercial adoption of its AI-powered precision medicine platform.

2. Tempus AI Reports Key 2025 Milestones

In a year marked by rapid growth, Tempus AI announced that its total contract value (TCV) reached $1.1 billion by the end of 2025, supported by over 70 new collaborations with pharmaceutical and biotech partners for data licensing and research services. The company also achieved its first positive adjusted EBITDA in the third quarter of 2025, reflecting improved operational leverage as volume-driven revenue scaled more rapidly than R&D and infrastructure investments. These milestones underscore Tempus AI’s progress toward sustainable profitability while expanding its footprint in both clinical and drug development markets.

3. Launch of Paige Predict Advances Digital Pathology

Tempus AI unveiled Paige Predict, a novel suite of AI-driven digital pathology tools that analyzes hematoxylin and eosin whole slide images to predict the presence of 123 clinically actionable biomarkers across 16 cancer types. Built on a foundation model trained on de-identified data from over 200,000 patients, Paige Predict delivers results directly to clinicians to inform sequencing and immunohistochemistry testing order. This capability addresses tissue scarcity challenges by guiding test prioritization, potentially reducing repeat biopsies and accelerating time-to-result for precision oncology workflows.

Sources

BZB