Ataraxis AI

• Institutions that require well-established assays with broad guideline endorsement and long clinical adoption histories may prefer established genomic tests for certain use cases.

• Organizations lacking digital pathology readiness (no reliable WSI generation, no scanner access, or insufficient PHI controls) are likely to face implementation challenges.

• Procurements that mandate publicly documented security certifications (SOC 2, ISO 27001) and cannot accept vendor-provided evidence on request may find the current public footprint insufficient.

• Organizations that require transparent, posted per-test pricing prior to engagement may prefer vendors that publish list prices and reimbursement pathways.

• Evidence maturity

  • Some comparative performance claims are supported by preprints rather than peer-reviewed publications; clinical utility and guideline incorporation remain to be established via independent replication and formal publication.

• Public documentation gaps

  • Limited public detail on security certifications (SOC 2 / ISO 27001), data residency, SSO/SCIM, audit logs, and technical integration capabilities may create procurement friction.

• Brand/name confusion on review sites

  • Some review listings for the name "Ataraxis" correspond to unrelated companies or services, which can complicate automated due diligence.

• Operational dependency

  • Requires a digital pathology pipeline; smaller clinics and practices without scanning and image transfer capability may face long lead times for implementation.

• AI-based multimodal prognostics using routine pathology

  • The product uses H&E morphology combined with clinical and clinicopathologic data to produce prognostic and predictive outputs without additional wet-lab testing.

• Turnaround and tissue preservation (company-stated)

  • The company states an approximate 24-hour turnaround and reports that the approach avoids consuming tissue required for other assays.

• Foundation model positioning (technical description)

  • The company describes a pan-cancer pathology foundation model trained on large-scale pathology data to support multimodal prognostication and downstream tasks.

Note: these points are primarily vendor-asserted and should be validated through peer-reviewed publications, independent replications, and operational pilots.

• On general SaaS review platforms, "Ataraxis" appears to correspond to an unrelated Workday consulting/services profile; this indicates a name collision on some review sites, which complicates automated sentiment aggregation for the oncology diagnostics product.

• G2 and similar review listings are likely not attributable to Ataraxis AI. No clearly attributable TrustRadius, Capterra, or Gartner Peer Insights entries for the oncology diagnostics product were identified in accessible searches.

• Unicancer partnership (Jan 5, 2026): retrospective analyses using data from a substantial patient cohort are reported as a validation activity for the breast prognostic tool. These announcements indicate evidence-generation efforts, but accessible materials did not include peer-reviewed outcome publications at the time of review.
• MEDSIR collaboration (Oct 2, 2025): an announced collaboration to evaluate the Ataraxis Breast tool in randomized clinical trial datasets and to investigate biomarkers for treatment optimization. Outcomes of these projects were pending in publicly available sources.

• Organization size

  • Mid-to-large health systems and academic medical centers with or planning digital pathology infrastructure are likely the best fits.

• Industries/verticals

  • Oncology care delivery, hospital pathology labs, cancer centers, and clinical trial sponsors seeking biomarker analyses.

• Roles / personas

  • Decision makers: CMIOs, pathology lab directors, molecular tumor board leaders, diagnostic lab leadership.
  • End users: medical oncologists, pathologists, tumor board coordinators, and care navigators.

• Pain points addressed

  • Uncertainty about recurrence risk and individualized treatment benefit, delays associated with some molecular tests, and limitations on available tissue for additional assays.

• Qualification criteria (good fit)

  • Existing access to digitized H&E slides or a near-term plan to digitize tissue workflows.
  • Institutional governance capable of evaluating AI solutions (clinical validation review, bias/fairness assessment, workflow integration planning).

• Deal disqualifiers (poor fit)

  • Limited or no digital pathology capability and no plan to implement within procurement timelines.
  • Procurement requirements that demand publicly posted pricing information or predefined payer coverage before purchase.

• What the product does (core value proposition)

  • Ataraxis AI develops machine-learning-based oncology prognostic and predictive tests that analyze routine H&E whole-slide images (WSI) combined with clinicopathologic and clinical variables to estimate outcomes such as recurrence risk, treatment benefit, and neoadjuvant response. The initial focus is on breast cancer under the Ataraxis Breast product line. (Company website and product pages)
  • The company describes a pan-cancer pathology foundation model (named "Kestrel") trained on a large corpus of pathology slides as a core technical asset.

• Target market and market positioning

  • Primary buyers and users appear to be health systems, oncologists, and pathology/diagnostic workflows seeking prognostic or predictive biomarkers that do not require additional tissue for molecular testing.
  • The company states the offering provides: (1) a short turnaround (company-stated target of roughly one day), (2) no additional tissue consumption compared with standard molecular tests, and (3) claims of potential improvements in prognostic accuracy relative to some standard tools (supported by vendor materials and preprints).

• Company background (founding year, HQ, key leadership)

  • Founded: commonly reported as 2022 (secondary sources).
  • HQ: New York, NY is consistently referenced in press and public coverage.
  • Key leaders/founders: Jan Witowski, MD, PhD (Co-founder/CEO) and Krzysztof Geras, PhD (Co-founder) are named in public announcements.

• Funding and growth (rounds, acquisitions, customer count/revenue if public)

  • Seed: reported to have completed an early seed financing round.
  • Series A: a Series A was announced in 2025 according to press coverage.
  • Total funding (in press): reported total funding in the low tens of millions; customer counts and revenue are not publicly disclosed in accessible sources.
  • Partnerships / validation collaborations: the company lists collaborations with cancer centers and academic groups; press materials reference collaborations focused on randomized clinical trial datasets and validation activities.

• Pricing model with specific tiers and costs where available

  • Public per-test pricing is not transparently published on accessible pages. The website includes an ordering/clinical tests flow, but no clear public list price was identified. As with many clinical diagnostics, final price may depend on reimbursement, contracting, and ordering channel.

• Total cost of ownership (TCO) considerations

  • Workflow and digitization prerequisites: Because the product relies on digital pathology WSI, TCO often includes scanner access or fees, digital storage, pathology IT integration, and change management.
  • Implementation / integration: A physician portal is referenced, suggesting an ordering and results workflow, but enterprise-grade integration details (e.g., HL7/FHIR, LIS connectors) are not fully documented publicly.
  • Clinical adoption cost: institutions will need to address clinician training on interpreting AI-derived risk/benefit metrics and incorporate results into tumor board and care pathways.

• Key competitors in the space (4–6 direct alternatives)

  • Exact Sciences (Oncotype DX) — genomic recurrence and treatment decision support
  • Agendia (MammaPrint/BluePrint) — genomic risk and subtyping
  • Foundation Medicine — genomic profiling (adjacent budget/clinical line)
  • ArteraAI — AI-driven predictive/prognostic tests in oncology (adjacent category)
  • Paige — computational pathology platforms (competes for digital pathology footprint)
  • PreciseDx (PreciseBreast) — digital pathology prognostic test positioned alongside genomic assays