Tempus AI: 83% Revenue Growth, 2% AI Revenue — When the Narrative Outpaces Reality
Spruce Point calls it AI-washing. Management calls it the next NVIDIA moment. Our 6-lens committee analysis found a genuinely defensible data moat wrapped in a narrative that may not survive sustained scrutiny.
+83% YoY (includes Ambry acquisition)
$12.4M of $693M in FY2024
Via 10b5-1 plans (Nov 2025-Mar 2026)
$1,640 vs >$2,200 target
Tempus AI may be one of the most interesting mismatches in public markets today: a company with a genuinely unique competitive asset (450+ petabytes of multimodal patient data), real revenue momentum (83% growth to $1.27B), and a data licensing business that pharmaceutical companies keep coming back to (126% net revenue retention) — all wrapped in an AI narrative that systematically overstates the company's current AI revenue contribution.
The company rebranded from “Tempus Labs” to “Tempus AI” before its June 2024 IPO. In FY2024, AI applications generated $12.4 million — roughly 2% of total revenue. The rest came from genomic diagnostic testing and data licensing, which are valuable businesses in their own right but are not what investors typically picture when they hear “AI company.”
Spruce Point Capital published a “Strong Sell” report in May 2025, alleging 50-60% downside and flagging AI-washing, inflated Total Contract Value, and CEO Eric Lefkofsky's historical pattern of post-IPO selling. The stock dropped 19%, a securities class action followed, and Tempus dismissed it all as “inaccuracies.” Then the business kept growing.
Want the full 6-lens analysis with signal assessments and model debates?
Opus + Sonnet ensemble. 6 lenses. 7 signals. 6 debates. Full evidence citations.
Signal Assessments
TCV includes $300M non-binding opt-ins and $200M related-party Pathos deal, inflating demand metric by ~27%
Dual-class voting control + 775K+ CEO shares sold via 10b5-1 while describing FY2026 as 'phenomenal'
Core organic growth ~33% is healthy; headline 83% includes acquisition. ASPs $500+ below target
FDA clearance pathway is #1 margin lever. Securities class action pending. MRD reimbursement uncertain
Trades at AI multiples (~7x revenue) but generates 98% of revenue from diagnostics and data licensing
~5.6x forward revenue includes AI premium that currently generates negligible revenue
450PB multimodal data, 5,500+ hospitals, 8,500+ oncologists create 5-10 year replication barrier
Key Findings
The Data Moat Is Real and Growing
Tempus's 450+ petabytes of multimodal patient data (molecular, imaging, clinical outcomes) spanning 500K+ records is the company's most valuable asset. With 5,500+ hospital connections and 8,500+ ordering oncologists, the clinical distribution network creates genuine switching costs. Data licensing grew 38-69% per quarter with 126% net revenue retention. Replicating this would take competitors 5-10 years and hundreds of millions of dollars.
TCV Metric Reliability Is Questionable
The $1.1B+ Total Contract Value that management uses to signal demand visibility includes $300M in non-binding opt-ins and a $200M Pathos deal involving entities founded and funded by company leadership. These components inflate TCV by approximately 27%, meaning forward revenue visibility may be weaker than quarterly presentations imply.
FDA Clearance Pathway Is the Primary Value Driver
Tempus's ASPs ($1,640) sit $500+ below peer levels and the company's own target of $2,200+. The single biggest lever for closing this gap is FDA clearance and ADLT migration for xT CDx, xF, and xR. Management plans to migrate the majority of xT CDx volume to the FDA-approved version by end 2026. Successful execution would transform unit economics; delays leave margins compressed indefinitely.
The AI Narrative Creates a Valuation Premium That May Not Survive Scrutiny
Tempus trades at ~5.6x forward revenue, reflecting AI-platform positioning. Pure diagnostics comps trade at 4-8x depending on growth profile. The premium above diagnostics peers reflects AI optionality from the foundation model, which currently generates zero revenue. CEO Lefkofsky references a potential “NVIDIA moment” while the company invested in 1,000+ H200 GPUs and 500+ GB200 GPUs. The investment is real; the revenue timeline remains undefined.
Where Models Disagreed
Is TCV a Meaningful Metric?
Opus Position
Non-binding opt-ins are standard in enterprise SaaS; $300M represents upside scenarios, not fabrication. TCV is directionally useful.
Sonnet Position
27% non-binding with a $200M related-party deal is materially higher than SaaS norms. TCV is unreliable as presented.
Resolution: TCV is useful directionally but the magnitude of non-binding components and Pathos related-party nature make it less reliable than management implies. Neither a smoking gun nor a clean bill of health.
Does the AI Rebrand Constitute Misrepresentation?
Opus Position
AI is embedded throughout genomics workflows. The rebrand reflects the platform's AI-driven capabilities, not standalone AI products.
Sonnet Position
Timing the rebrand to the IPO with 2% AI revenue is opportunistic. Investors reasonably interpret “AI company” as primary AI-driven revenue.
Resolution: The rebrand is aggressive marketing but defensible. AI is embedded in workflows even though standalone revenue is negligible. Investor expectations may systematically overstate the AI contribution.
Can Incumbents Replicate the Data Moat?
Opus Position
5-10 year replication timeline makes the moat practically impregnable in the medium term.
Sonnet Position
Roche owns both Flatiron Health (data) and Foundation Medicine (diagnostics). Integration could challenge Tempus within 2-3 years.
Resolution: Replication is theoretically possible but practically unlikely in the medium term. The realistic threat is horizontal expansion by specialists, not full-stack replication.
Cross-Lens Reinforcements
Governance-Narrative Alignment Concern (3 lenses)
Fugazi Filter, Insider Investigator, and Myth Meter independently flagged the same pattern: structural protection from accountability combined with promotional tone that exceeds operational reality.
FDA Pathway Is the Primary Value Driver (2 lenses)
Regulatory Reader and Gravy Gauge both identified FDA clearance as the single biggest lever for ASP improvement ($500+ per test) and margin expansion.
Data Moat Real, AI Premium Questionable (2 lenses)
Moat Mapper validated the data repository as defensible. Myth Meter found the AI narrative inflates expectations beyond what the data asset alone justifies.
What to Watch
Management targets majority of xT CDx volume on FDA-approved version by end 2026. Below 50% by Q3 2026 signals execution issues that would delay the $500+ ASP improvement.
First $50M+ quarter from AI applications would validate the AI narrative and justify the valuation premium. Continued near-zero AI revenue strengthens the re-rating risk.
Monitor the non-binding opt-in percentage and related-party deal proportion when Tempus provides annual TCV detail. A rising opt-in percentage signals metric degradation.
If the AI-washing class action survives the motion to dismiss, discovery could surface internal communications about the timing and intent of the Tempus Labs to Tempus AI rebrand.
HIGHER SCRUTINY
Tempus AI has a genuinely defensible data moat and strong revenue momentum, but governance structures and AI narrative inflation create information asymmetry that warrants elevated scrutiny. The gap between operational reality (a diagnostic testing and data licensing company growing ~33% organically) and market positioning (an AI platform trading at ~5.6x forward revenue) means downside is amplified if the market re-rates to diagnostics multiples, even with strong execution.
Path to More Favorable Assessment
- • First meaningful AI algorithm revenue ($50M+ quarter)
- • ASP reaching $2,000+ through FDA clearance pathway
- • CEO selling pace reduction or insider buying
- • Securities class action dismissal
Path to Less Favorable Assessment
- • TCV non-binding opt-in percentage rising
- • Securities class action surviving motion to dismiss
- • Organic growth decelerating below 25%
- • Foundation model delays beyond 2027
This analysis is for educational purposes only — it is not a recommendation to buy or sell any security.
Public Sources Used (13 documents)
- • Annual Report (10-K) — FY2025
- • Quarterly Reports (10-Q) — Q3, Q2, Q1 2025, Q3 2024
- • Current Reports (8-K) — 10 filings (2025-2026)
- • Proxy Filing (DEFA14A) — April 2025
- • Form 4 Insider Transaction Filings — 20 filings
- • Q4, Q3, Q2, Q1 2025 Earnings Call Transcripts
- • Spruce Point Capital — Strong Sell Report on Tempus AI (May 2025)
Full Analysis with Signal Breakdowns
Explore the complete 6-lens assessment including debate transcripts, evidence citations, and monitoring triggers.
View TEM Analysis