1. What happened
An October 22, 2026 order form licenses the records and Tempus Agent to Pathos for internal use and to assist BMS; a side letter expressly permits Pathos to sublicense the data and agent to BMS.
2. Buyer / seller
Pathos AI is the AI buyer or model-development side. Tempus AI owns or contributes the proprietary corpus.
3. Economics
Up to $55 million: $50 million at $100 per delivered record plus $5 million for availability of record updates; realized cash receipts depend on delivery and are not verified. Economics scope: The filing separately identifies record-license fees and update availability; no standalone value is assigned to the Tempus Agent.. Strategic Signal has not estimated undisclosed consideration.
4. Proprietary data
Up to 500,000 de-identified patient records covering Alzheimer's disease and other specified neurological conditions, targeted for delivery at 100,000 records per year over five years with update availability. Historical depth: not_disclosed. Refreshability: Targeted annual delivery over five years plus separately paid availability of record updates..
5. Rights and restrictions
Use is governed by the Neuro Order Form and Side Letter for Pathos internal purposes and BMS therapeutic-development support. Ownership: No ownership transfer is disclosed.. Training: not disclosed. Post-training: not disclosed.
6. Why the data matters for AI
Interrogation and analysis of neurology records through the Tempus Agent and Pathos Foundry to support therapeutic development; model training is not disclosed. The filing does not identify a decision-to-outcome field structure.
7. Asset Score — 92/100
A specified five-year neurology corpus with annual delivery, record updates and governed analysis access is highly valuable; historical depth and outcome-field structure remain undisclosed. Score coverage: 79%.
| Factor | Weight | Rating / 5 | Points |
|---|---|---|---|
| Uniqueness / scarcity | 12 | 4.5 | 10.8 |
| Historical depth | 8 | Unknown | — |
| Decision → outcome richness | 13 | Unknown | — |
| Decision Value Density | 13 | 4.5 | 11.7 |
| Real-world grounding | 10 | 5 | 10.0 |
| Domain economic value | 9 | 5 | 9.0 |
| Refreshability | 8 | 5 | 8.0 |
| Proprietary advantage | 9 | 4.5 | 8.1 |
| Model-learning usefulness | 10 | 4 | 8.0 |
| Non-replicability | 4 | 4 | 3.2 |
| Rights usability | 4 | 4.5 | 3.6 |
8. Transaction Signal Score — 85/100
The filing discloses per-record economics, delivery volume, updates and BMS sublicensing, but realized delivery, competing bids and independent model uplift remain unknown. Score coverage: 85%.
| Factor | Weight | Rating / 5 | Points |
|---|---|---|---|
| Disclosed economics | 16 | 4.5 | 14.4 |
| Clean price discovery | 16 | 3.5 | 11.2 |
| Separability of data consideration | 15 | 5 | 15.0 |
| Explicit training / model-improvement use | 14 | 3.5 | 9.8 |
| Clarity of rights purchased | 12 | 4.5 | 10.8 |
| Competing bids | 10 | Unknown | — |
| Strategic buyer quality | 6 | 4 | 4.8 |
| Repeatability / precedent value | 6 | 5 | 6.0 |
| Independent model-uplift evidence | 5 | Unknown | — |
9. M&A / valuation implications
The per-record fee and update component provide a transparent licensing precedent, while delivery risk and bundled agent access limit clean corpus-only comparability.
10. Evidence boundary
This record covers the Tempus-to-Pathos neurology license and disclosed BMS sublicense. It is separate from the Tempus-to-BMS oncology license. Training rights, realized deliveries, competing bids, and independent model uplift are not disclosed.