Problem

Therapeutic antibody discovery is slow, expensive, and inefficient: researchers must search through billions of possible antibody sequences, yet conventional approaches often fail to identify antibodies that are both functional and therapeutically useful. The standard process separates hit discovery from lead optimization, with roughly nine months spent finding hits and another 18 months across 20–30 optimization cycles. The economic pain is especially acute for difficult, poorly characterized, or previously undruggable targets, where the available experimental data and screening approaches are weakest.

DenovAI’s central use case is to make high-affinity antibodies and small-protein binders for selected targets—including challenging targets in oncology, immunology, and inflammatory disease—without relying on the huge screening campaigns and iterative refinement traditionally required. Its intended outcome is to expand the range of diseases and targets that can yield viable biologic therapies while reducing discovery time and cost.

Product / Service

DenovAI is developing an AI–physics asset-discovery engine that designs therapeutic proteins de novo, entirely from scratch. Its platform combines generative AI, molecular biophysics, protein-structure prediction, molecular dynamics, and automated workflows, integrating sequence, structure, kinetics, and function to generate and optimize antibodies, miniproteins, multispecifics, and other protein modalities. Unlike approaches that begin with known binders or large training datasets, DenovAI says its system can design candidates without a starting template and can narrow experimental testing to a small number of candidates rather than thousands.

The company is pursuing a hybrid model: it is building an internal pipeline of proprietary therapeutic assets while selectively working with pharmaceutical companies through co-discovery and milestone-based collaborations, rather than relying solely on software licensing. DenovAI says this “one-shot” design approach builds desired properties into molecules before wet-lab testing; its public materials claim the potential to compress discovery from months or years to days or weeks, while the company’s 2026 business description says it is advancing selected programs toward IND-enabling work.

Market

DenovAI competes in the AI-enabled biologics-discovery and computational protein-design market, specifically the emerging segment focused on de novo therapeutic antibodies and protein biologics. The market is attracting substantial pharmaceutical interest because AI platforms can combine computational design with biologics expertise to accelerate antibody discovery. Relevant adjacent or overlapping competitors include Absci and BigHat Biosciences in AI-driven antibody design, Generate:Biomedicines and Xaira in generative biologics design, and BioMap in broader AI-enabled biologic design. DenovAI’s positioning is differentiated by its emphasis on de novo design integrated with molecular biophysics and dynamics, rather than primarily iterating from existing binders.

DenovAI is a development-stage biotech rather than an established commercial drug company. Founded in 2023 as an AION Labs portfolio company, it has received seed funding and backing connected to AstraZeneca, Merck, Pfizer, Teva, AWS, the Israel Biotech Fund, Amiti Ventures, the Israel Innovation Authority, and BioMed X. Its technical traction includes a peer-reviewed AlphaDesign study in which 17 of 88 de novo protein designs showed in-vivo activity, while later materials describe discovery programs and a pipeline spanning oncology, immunology, and inflammation. The available evidence does not disclose product revenue or clinical-stage assets; instead, it describes near-term revenue as expected from milestone-based pharmaceutical collaborations while the company develops proprietary programs and prepares for a Series A round.