PUBLIC RESEARCH · WEEKLY
Recursion Adds Human Efficacy Signals and Tempus Adds Revenue Proof, but Sample Size and Cash Remain Hard Gates
AI Biotech Weekly | Issue 013 | 4 November 2026
Recursion expands Phase 2 REC-4881 FAP data and reports early REC-1245 tumor responses, while Tempus Q3 revenue reaches $411.5m, up 23%, with about 12,000 MRD tests, up roughly 300%, and a higher full-year outlook. The ACIS score rises from 95 to 96: human and commercial evidence improve together, but Recursion's samples remain small and Tempus still reports a $135.1m nine-month net loss with substantial SBC.
Listed securities referenced: Recursion Pharmaceuticals, Tempus AI, Sanofi, Merck and Moderna.
Recursion expands Phase 2 REC-4881 FAP data and reports early REC-1245 tumor responses, while Tempus Q3 revenue reaches $411.5m, up 23%, with about 12,000 MRD tests, up roughly 300%, and a higher full-year outlook. The ACIS score rises from 95 to 96: human and commercial evidence improve together, but Recursion's samples remain small and Tempus still reports a $135.1m nine-month net loss with substantial SBC.
The one-minute brief
Platform commercialization and human clinical validation form a clearer dual evidence chain. Recursion now has polyp-burden and early tumor-response signals, while Tempus turns diagnostic volumes and data applications into quarterly revenue. Neither chain should be overstated: open-label small samples and short follow-up do not establish approval probability, while revenue and adjusted EBITDA do not replace CFO, FCF and dilution analysis. The thesis upgrades from payment validation to clinical signal plus commercial execution, with Decision State still subject to Human Review.
Audio transcript
The AI biotech score rises from ninety-five to ninety-six. Recursion adds stronger human signals. REC-4881 reduces polyp burden in familial adenomatous polyposis, while REC-1245 shows two confirmed partial responses among twelve evaluable patients. Tempus adds commercial execution, with quarterly revenue of four hundred eleven point five million dollars and MRD volume up about three hundred percent. Small samples, open-label data, nine-month losses and high stock compensation remain hard boundaries. Clinical and revenue evidence are not approval or free cash flow, and investment action still requires human approval.
Prior 95 | Change +1 | Human and revenue evidence improve together
REC-4881 expanded FAP data; very early REC-1245 responses
Tempus revenue $411.5m; MRD volume about +300%
Recursion runway extends; Tempus loss and SBC remain high
Human Review; no automatic eligibility upgrade
01 · SCORE & STAGE
Why does clinical plus commercial validation add only one point?
| Dimension | Prior | Current | ACIS read |
|---|---|---|---|
| Composite score | 95 | 96 | Human and revenue evidence improve; replication, approval and cash remain hard gates. |
| Scientific validation | 97 | 98 | Multiple clinical programs and partner actions support platform output. |
| Clinical validation | 91 | 94 | Polyp burden and early tumor responses emerge with sample and control limits. |
| Pharma / partner validation | 100 | 100 | Sanofi exercises its first candidate option and Tempus adds Merck/Moderna work. |
| Pipeline quality | 95 | 97 | REC-4881, REC-1245 and the Sanofi candidate advance. |
| Cash runway / financing | 98 | 98 | Recursion runway extends into 2029; dilution and burn remain. |
| Commercialization / economics | 98 | 99 | Tempus revenue and test volumes deliver; CFO and FCF quality remain open. |
| Valuation / expectations | 79 | 77 | Better news can raise expectations faster than safety. |
| Regulatory / execution risk | 86 | 87 | Execution improves while clinical, privacy, reimbursement and cash risks remain. |
Evidence moves in the same direction, but REC-4881 and REC-1245 remain small, short or open-label. Tempus growth coexists with a $135.1m nine-month net loss and $154.9m of SBC; FCF quality is not closed.
02 · EVIDENCE MAP
Four signals keep human, partner, revenue and cash evidence separate
| Signal | Confirmed evidence | Investment read | Evidence boundary |
|---|---|---|---|
| REC-4881 FAP data | At 12 weeks, nearly half of evaluable upper-GI and more than one-third of lower-GI patients achieved at least 65% polyp-burden reductions; 85% remained below baseline 12 weeks off treatment. | Human lesion burden shows stronger directional and persistence signals. | Open-label small samples and short follow-up do not replace randomized or long-term outcomes. |
| REC-1245 early responses | Among 12 evaluable patients at at least 45mg, two confirmed and one unconfirmed partial responses were reported, with 91.7% disease control. | A platform-derived asset shows early anti-tumor activity. | The sample is tiny and heterogeneous and cannot imply registration success. |
| Sanofi option | Sanofi exercises the first candidate option under the collaboration, with Phase 1 expected in 2027. | Pharma continues to pay for platform output. | An option and milestones are not clinical success, revenue or FCF. |
| Tempus commercial execution | Q3 revenue was $411.5m, up 23%; Diagnostics was $303.5m, Data & Applications $108.1m, and MRD volume about 12,000, up roughly 300%. | Diagnostics and data move from contracts into recognized revenue. | Nine-month net loss was $135.1m and SBC $154.9m; revenue is not owner cash. |
03 · RESEARCH PRIORITY
The view first: clinical and commercial gates improve while capital return remains unproven
| Company | Current view | Proof required | First rejection risk |
|---|---|---|---|
| Recursion / RXRX | Clinical signal up / still early | Larger controlled cohorts, durability, safety, regulatory path, cash use and dilution. | Signals fail to replicate or financing dilutes. |
| Tempus / TEM | Revenue and volume deliver / cash under audit | Gross margin, CFO, FCF, SBC, receivables, reimbursement and renewals. | Growth fails to become cash. |
| Sanofi / SNY | Platform validation improves / return UNKNOWN | Clinical entry, milestones, shared cost and asset value. | Option does not translate into success. |
| Merck / Moderna | Partner breadth expands / attribution limited | Revenue, delivery, use, renewals and clinical efficiency. | Headlines precede attributable outcomes. |
Research priority is not a buy, sell or position-size recommendation.
04 · EVIDENCE LADDER
From platform to drug cash: human and revenue gates improve; approval and FCF do not clear
| Evidence layer | Weekly state | ACIS read |
|---|---|---|
| Data / model | Continued validation | Platform output receives pharma options and clinical readouts. |
| Candidate / wet lab | Up | REC-1245 and the Sanofi candidate move model output into development. |
| Human safety / efficacy | Up but early | Directional FAP and oncology signals need larger controlled data. |
| Regulatory / approval | Not cleared | No registrational result or approval is added. |
| Diagnostics / data revenue | Green / revenue confirmed | Tempus test volume and data revenue are auditable. |
| CFO / FCF | Orange | Loss, SBC, burn and financing keep cash quality open. |
| Capital markets | Runway extended / dilution present | Recursion financing lowers near-term liquidity risk but adds dilution. |
05 · INVESTMENT IMPLICATIONS
Where does evidence quality improve?
Dual evidence beats a single narrative
Clinical signal and revenue improve platform credibility.
Human signals require replication
Small open-label data do not replace controlled outcomes.
Revenue is not cash
Loss, SBC, CFO and FCF remain separate.
Human Review remains required
Evidence updates do not automatically change positions.
06 · PRINCIPAL RISKS
What can break the commercialization chain?
Small samples
Early response rates may fall in larger cohorts.
Clinical and regulatory
Safety, durability and approval remain uncertain.
Cash and dilution
R&D and financing may dilute returns.
SBC and accounting
Adjusted profit can overstate owner cash.
Data and reimbursement
Privacy, payers and collection may constrain growth.
07 · NEXT VALIDATION
What matters over the next 90 days?
| Window | Evidence to watch | Warning signal |
|---|---|---|
| Next 7 days | Recursion and Tempus 10-Q detail on patients, cash, SBC, receivables, CFO and capex. | Percentages replace absolute patient counts. |
| Next 30 days | REC-4881/1245 follow-up, safety and expansion; Tempus margin and cash conversion. | Signals or revenue fail to replicate. |
| Next 90 days | Sanofi candidate IND work, Tempus delivery, CFO, FCF and reimbursement quality. | Partnerships and financing rise without outcomes or cash. |
Upgrade conditions
Upgrade if REC-4881 and REC-1245 replicate efficacy and safety in larger comparable cohorts and Tempus converts revenue into sustainable CFO and FCF with lower dilution and SBC.
Downgrade conditions
Downgrade if clinical signals weaken, regulatory timing slips, Tempus growth fails to convert into cash, or financing and SBC dilute shareholder value.
FAQ
Key questions
Do REC-4881 data prove efficacy?
They are directional human evidence, not registrational proof.
Can the 91.7% REC-1245 disease-control rate be extrapolated?
No. Only 12 patients were evaluable.
Does Tempus revenue growth prove an FCF inflection?
No. Loss, SBC, receivables, CFO and capex still require audit.
Does the Sanofi option mean drug success?
No. It is a development commitment, not approval.
Is this investment or medical advice?
No. Weekly updates Evidence; decisions require professionals and Human Approval.
KEY TERMS
Reader reference
Statement of Work / SOW — A contract describing scope, term, payment and performance.
Fixed Payments — Scheduled consideration not directly tied to each record.
Per-record Pricing — Fees based on records actually delivered.
De-identified Data — Data processed to remove or reduce personal identification under applicable rules.
Related Person Transaction — A transaction requiring added conflict review because of management, board or ownership ties.
Agentic Tool — Software that performs multi-step analysis inside authorized data and workflows.
SOURCES & CONTINUITY
Primary sources and evidence boundaries
Recursion | Third-quarter 2026 results and clinical updates | 2 November 2026 ↗
Recursion | REC-4881 and REC-1245 presentations | 2 November 2026 ↗
Related ACIS research
Evidence is current through 4 November 2026, 17:08 Beijing time. Recursion and Tempus 2 November releases, presentations and 10-Q are Tier-A corporate and regulatory sources. Clinical percentages are company-reported and their small, open-label or short-follow-up boundaries are explicit. Tempus nine-month loss, SBC and revenue come from the 10-Q; fields without complete same-basis verification remain UNKNOWN. This evidence is Earnings Intelligence L3 and is shared with CPW and Investment Intelligence for Human Review without automatically changing Decision State. For research and education only; not investment or medical advice.