On September 9, Tempus AI (NASDAQ:TEM) said ARPA-H, a US health research agency, picked it to build an AI that treats heart failure patients without a clinician stepping in. The award is worth up to $9.5 million, which is small change for a company booking hundreds of millions in quarterly sales. But the dollars are not really the story. The story is where Tempus wants to point its data and software next.
Heart Failure’s Missing Doctors
Heart failure is a problem of access as much as medicine. It affects more than 6 million Americans, and nearly half do not survive five years past diagnosis, even though life-extending treatments exist. Under 25% of patients get every recommended therapy, and nearly 46% of US counties have no practicing cardiologist. Tempus is betting software can stretch that thin workforce. Its agent would watch hospital records and home heart rate monitors around the clock, then talk to patients directly about symptoms, medications, virtual rehab and scheduling, and write everything back into the medical record.
Tempus also is not starting from a blank page. Its care-gap tool, Tempus Next, has combed through more than 2.7 million cardiology patients, and the company holds three FDA-cleared ECG products that hunt for undiagnosed heart disease. Meanwhile, the core business has momentum. For the quarter ended June 30, reported on July 30, revenue rose 22% to $382.5 million, led by oncology testing volume up 31%. Data licensing grew faster still, up 36%, so drugmakers keep paying for what Tempus collects. With $820.7 million in cash and marketable securities, the company can chase cardiology without leaning on the grant.
Ambition Meets Fine Print
Start with the word “up” in up to $9.5 million. It is a ceiling, and the money pays for building the agent and testing it in a multi-center prospective study, not for selling it. Nothing announced so far shows the agent works. Letting software run cardiovascular care without direct human intervention is a high bar, and the trial still has to show the agent is safe and effective. Until it does, this is a plan rather than a product.
The financials call for some care too. Tempus posted net income of $5.6 million in the June quarter, but that leaned on $98.5 million in unrealized gains on marketable securities. Loss from operations actually widened 23% from a year earlier. Add the Personalis acquisition, announced July 20 and valued at about $1.5 billion in enterprise value, and you have a company piling big commitments onto a thin profit picture. The deal is expected to close in late Q4 2026 or early 2027, and guidance leaves it out entirely.
Buyers and Shorts Collide
Hedge fund ownership climbed to 38 funds from 33 in the latest quarter, so more big-money holders are getting in. But 29.17% of the float is sold short, which is heavy skepticism, and it means a lot of money is wagering that the growth story falls short. Crowded shorts can fuel a sharp squeeze if the news keeps landing well, though some of that positioning may simply be hedging.
Proof Still Has to Arrive
The ARPA-H award is a small check attached to a very large idea, and that gap is the whole debate. Tempus has a growing core business and a real head start in cardiology data, but an autonomous heart failure agent is still a promise until the prospective study reports. Bulls need that trial to show safe, effective care while oncology and data licensing keep paying the bills. Bears are watching whether widening operating losses and a big acquisition crowd out the experiment.
READ NEXT: 10 Best Future Stocks to Buy Under $10 and 12 Best Performing Semiconductor Stocks to Invest In.
