The gap between a promising objective and an existing product
Theranos proposed an appealing change to laboratory testing: extensive, fast and affordable tests using very small blood samples. Such a goal was not itself fraudulent. The case concerned what investors and patients were told the company could already do, what executives knew about limitations, and how those representations influenced decisions. The distinction between an aspiration and a demonstrated capability is central to understanding the business history without equating every failed invention with a crime.
The SEC's March 2018 case alleged that Theranos, founder Elizabeth Holmes and former president Ramesh “Sunny” Balwani raised more than $700 million through misleading claims about technology, business and financial performance. The agency said the proprietary analyzer performed only a small number of tests while most patient tests used other manufacturers' equipment. Theranos and Holmes settled the civil charges without admitting or denying the allegations. Later criminal trials established narrower, defendant-specific verdicts that must be described separately. [1]
The technology claim had several layers
A blood-testing service involves collecting a sample, preparing it, measuring the relevant substance, maintaining quality controls and reporting a result. Success at one stage does not establish success at every stage. A small collection container is not proof that an analyzer can reliably perform every advertised test; a working test is not proof that the same device supports an entire testing menu. These distinctions matter to both the medical service and the economics of scaling it.
The Ninth Circuit's account describes Theranos's Edison as the company's proprietary device actually used for patient testing, while other devices remained in development. It states that only twelve assays were run on the Edison and describes the use of third-party machines, including modified equipment. An assay is a particular analytical test. The fact that tests were available through a Theranos-branded service therefore did not establish that its proprietary platform could perform them. [2]
Using commercial equipment is not inherently improper. The significance here was the divergence between the process used and the process represented. A service built around established analyzers can be useful, but its cost structure, technological differentiation and expansion prospects differ from those of a new general-purpose testing platform. Investors assessing those two businesses would be assessing different technical and competitive propositions even if the customer-facing brand were identical.
Demonstrations, endorsements and military deployment
The SEC complaint alleged that certain demonstrations led visitors to believe samples were being processed on Theranos equipment even when samples were processed elsewhere. It also described misleading materials bearing pharmaceutical-company logos and claims of outside validation. These allegations concerned the provenance of evidence, not merely whether an executive was optimistic. A demonstration is persuasive because observers infer that the displayed process generated the displayed result; that inference fails if the process is materially different. [3]
In its account of the trial evidence, the Justice Department said Holmes falsely represented that major pharmaceutical companies had comprehensively validated the technology, and that Theranos had a profitable military business with battlefield deployment. It also described claims of a planned Walgreens expansion from a few dozen locations to 900 despite a stagnating relationship. These representations supplied apparent confirmation from institutions outside Theranos. They were not interchangeable with the underlying institutions actually endorsing the claimed capabilities or commercial trajectory. [4]
The financial mechanism is straightforward: a credible outside customer can appear to resolve uncertainty that investors cannot directly evaluate. Yet a research discussion, pilot, commercial contract and broad deployment are different milestones. A company name on a presentation can convey more certainty than the actual relationship supports. The Theranos record illustrates how technical and commercial claims can amplify one another without providing genuinely independent corroboration.
Revenue projections were tied to present-tense representations
The SEC complaint alleged that investors were told Theranos had generated or would generate more than $100 million in 2014 revenue and was on a path to approximately $1 billion in 2015. It contrasted these claims with operating revenue of little more than $100,000 in 2014. Those figures concern the complaint's specified revenue measures and periods; they are not the company's valuation or the total capital it raised. [3]
Forecasts inevitably involve uncertainty. A forecast becomes materially different when its apparent foundation includes false statements about existing capabilities, completed validation, signed commercial expansion or current financial performance. A wide forecasting error alone cannot establish intent to defraud. Conversely, labeling a number a projection does not make its supporting factual representations unimportant. The history is therefore more precise when it examines what was represented as already achieved rather than treating the difference between a forecast and an outcome as the whole case.
In a hypothetical testing business, doubling locations produces little revenue if each location cannot perform the promised menu reliably or attract paying volume. Revenue depends on functioning operations, utilization, pricing and collection, not simply the number of prospective outlets. Capital raised can finance an operating loss, but cannot transform that loss into customer revenue. Theranos's claims joined a technical breakthrough to rapid distribution, making the evidentiary weakness in either component significant for the whole growth narrative.
Governance and the limits of borrowed credibility
Holmes combined founder and chief-executive authority with voting control. The SEC settlement included relinquishment of that control, return of shares, a $500,000 penalty and a ten-year public-company officer-and-director bar. These remedies addressed control and incentives as well as money. The settlement's existence did not mean that the SEC allegations had been admitted. [1]
The board question is not whether prominence is inherently suspect. Directors with political, military or commercial experience can contribute relevant judgment. But reputation in one field does not validate a laboratory's performance, and a distinguished name cannot substitute for evidence about what a device actually does. The analytical failure occurs when affiliation is treated as verification, especially if multiple participants are relying on the same management-supplied claims.
Likewise, investor sophistication does not make false information true or excuse fraud. Diligence and truthful disclosure are complementary responsibilities. An investor may misunderstand evidence, ask too few questions or overvalue a prominent endorsement; none of those possibilities gives an issuer permission to misrepresent facts. The case is thus not adequately explained by describing its investors as either unusually clever or unusually credulous. The relevant issue is the quality and independence of information available to them.
Clinical regulation was a separate evidentiary track
The FDA's September 2015 inspection form for Theranos's Palo Alto facility included observations about design validation, design-risk analysis and documentation. The form expressly cautions that inspectional observations are not a final agency determination of compliance. That qualification belongs with the source. A Form 483 is not a criminal verdict, nor does the existence of an observation establish that every test or every patient result was wrong. [5]
These distinctions prevent two opposite errors. Treating a regulatory inspection as conclusive proof of every criminal allegation overstates its role. Treating an investor-fraud verdict as the only relevant evidence of product quality understates the separate importance of laboratory oversight. Clinical reliability, compliance with a regulatory requirement and criminal intent are different questions, even when some of the same documents bear on each.
The financial consequences also extend beyond a faulty product's immediate sales. A laboratory's ability to operate, the reliability of its prior results and the credibility of its brand can affect future contracts and enterprise value. That is an analytical connection, not a calculation of damages or a judgment about an individual patient's medical outcome. No private patient circumstances are necessary to explain it.
The criminal verdicts were different
On January 3, 2022, Holmes was convicted of one investor-fraud conspiracy count and three investor wire-fraud counts. She was acquitted of the patient-related conspiracy count and three individual patient-fraud counts; the jury did not reach verdicts on three other investor counts. Another patient count had been dismissed during trial. On November 18, 2022, she received a 135-month prison sentence, plus supervised release. These details make “convicted of defrauding patients” an inaccurate description of her verdict. [4]
Balwani was tried separately. His jury convicted him on all twelve counts in July 2022, encompassing both investor and patient schemes. He was sentenced in December 2022 to 155 months, or twelve years and eleven months, with supervised release. The Justice Department's account describes evidence of inaccurate testing and representations to paying patients, as well as the investor deception. His result cannot be substituted for Holmes's, even though the trials concerned overlapping business conduct. [6]
An acquittal means the government did not establish the charged offense beyond a reasonable doubt. It does not certify that a product was reliable. A conviction on an investor scheme also does not automatically establish every possible patient injury or every allegation in earlier civil litigation. Respecting the different verdicts allows the history to explain both the commercial deception and the medical setting without using one legal finding to prove unrelated propositions.
Appeals, restitution and the dated status
The Ninth Circuit affirmed the convictions, sentences and restitution order in February 2025 and issued an amended opinion in December 2025. Its analysis included evidentiary challenges and the treatment of investor losses. It concluded that certain challenged testimony crossed into expert territory but that the errors were harmless. An appellate affirmance need not mean every trial ruling was flawless; it means the identified issues did not warrant the requested reversal under the governing standards. [2]
The HHS Departmental Appeals Board's January 2026 decision records joint-and-several restitution of $452,047,268 and upheld Holmes's ninety-year exclusion from federal health-care programs. That administrative exclusion is distinct from imprisonment, the SEC's public-company bar and an actual cash recovery for investors. A restitution obligation does not by itself establish how much money victims have received. [7]
As checked October 4, 2026, Balwani's Supreme Court docket showed a certiorari petition filed May 21, 2026 and distributed for the September 28 conference, with no disposition displayed on the retrieved docket. The petition concerns Balwani; it is not described here as Holmes's petition. A request for Supreme Court review is not a grant of review or a reversal. This article reports the verified docket status rather than assuming the conference produced a particular outcome. [8]
The lasting distinction
Theranos shows how a financing narrative can turn uncertain technical development into apparently established commercial achievement. Demonstrations, outside names, projected revenue and retail distribution seemed to reinforce one another. The central problem was that their apparent independence exceeded the evidence behind them. The resulting business story could attract capital without resolving the technical questions on which its economics depended.
The case does not condemn ambitious research or every company that misses a forecast. It distinguishes the possibility of future success from truthful statements about the present. Preserving that distinction, together with the separate civil, criminal, regulatory and appellate records, explains why Theranos was more than an unsuccessful startup while avoiding claims that the courts never actually decided.
Sources
- SEC, Theranos civil charges and settlement terms, March 14, 2018Filing / reportBack to text: ↑1↑2
- Ninth Circuit, amended opinion December 22, 2025, in Supreme Court-filed appendixOfficial source · PDFBack to text: ↑1↑2
- SEC, complaint against Theranos and Holmes, March 14, 2018; allegationsFiling / report · PDFBack to text: ↑1↑2
- DOJ, Holmes trial evidence, verdict and sentencing, November 18, 2022Official sourceBack to text: ↑1↑2
- FDA, Theranos Palo Alto inspection observations, September 16, 2015; not final compliance determinationOfficial sourceBack to text: ↑
- DOJ, Balwani verdict and sentence, December 7, 2022Official sourceBack to text: ↑
- HHS Departmental Appeals Board, Elizabeth A. Holmes, DAB 3222, 2026Official sourceBack to text: ↑
- Supreme Court, Balwani v. United States, No. 25-1330; docket checked October 4, 2026Official sourceBack to text: ↑