Did Yampolskiy's Alleged Competitor Hacking Cross the Line Drawn by the Supreme Court in Van Buren?
How the Supreme Court’s landmark CFAA decision applies to alleged unauthorized competitor access described in Factor Cybersecurity v. SecurityScorecard, and why valid credentials do not necessarily establish lawful authorization.
13 min readAnalysis · Not legal advice
Figure 1 · Two access scenarios, side by side
Both scenarios are illustrations of legal categories, not findings about anyone. Neither is dispositive: each still depends on facts a court would have to find.
Scenario A
An authorized customer retrieves data it is permitted to see, then shares it
A paying subscriber logs in with its own credentials, opens pages and reports the subscription entitles it to open, and later passes what it saw to someone else. The login is the subscriber's. The data sits behind a gate the subscriber was given a key to.
▸Credential belongs to the person using it
▸Every area opened was within the account's entitlement
▸The wrong, if any, is what happened to the data afterwards
Under Van Buren, retrieving information the account is entitled to retrieve is not “exceeding authorized access,” even if the purpose breaks a contract or policy. Downstream misuse is a contract or trade-secret question, not a §1030 access question.
Scenario B
An outsider uses a third party's credentials without the platform operator's consent
A competitor obtains login credentials issued to someone else and uses them to enter a platform it was never granted access to. The account holder may have handed the credentials over willingly. The platform operator never agreed to let this user in at all.
▸Credential belongs to someone else
▸The user has no account and no entitlement of its own
▸The operator, not the account holder, controls who may enter
This is the “without authorization” branch, which Van Buren did not decide. Courts ask whether the system owner gave this user permission. A technically valid login does not answer that question by itself, and a borrowed password is not automatically a crime either.
01The question Van Buren answers, and the one it leaves open
In 2021 the Supreme Court decided Van Buren v. United States, the most consequential interpretation of the Computer Fraud and Abuse Act in the statute's history. A police sergeant with lawful access to a license-plate database ran a search for money, in plain violation of department policy. The government said that made him someone who “exceeds authorized access.” The Court disagreed. An individual exceeds authorized access, it held, when he obtains information located in areas of a computer that are off limits to him, not when he obtains information he is entitled to obtain for an improper reason.
The decision narrowed the CFAA dramatically. Breaking a terms-of-service agreement, violating an employer's acceptable-use policy, or lying about why you are looking something up no longer converts ordinary use into a federal computer crime. The Court was explicit about the stakes: the government's reading would have criminalized a breathtaking amount of everyday activity.
What Van Buren did not resolve is the other half of the statute. Section 1030 reaches both access “without authorization” and access that “exceeds authorized access.” Van Buren was an exceeds-authorized-access case: the sergeant unquestionably had an account. The Court expressly declined to decide whether authorization turns only on technical barriers such as passwords, or also on words-based limits imposed by the system owner. That unresolved question is exactly the one raised by the allegations made in Factor Cybersecurity v. SecurityScorecard, where the asserted access was not by a subscriber exceeding its own entitlements but by an outsider alleged to have entered a competing cybersecurity ratings platform using credentials issued to someone else.
02Two scenarios, side by side
The infographic above sets out the distinction that drives everything else. In Scenario A, the person at the keyboard holds the account. Whatever they do with what they find, they were inside a gate the operator opened for them, and Van Buren takes that outside the CFAA's access prohibition. In Scenario B, the person at the keyboard holds someone else's credential and has no relationship with the platform operator at all. Authorization in that scenario is not established by the fact that the password worked.
Neither column is dispositive. Scenario A can still involve contract breach or trade-secret liability. Scenario B can still fail: if the operator knowingly tolerated shared logins, or if the account holder had the operator's blessing to bring others in, the outsider may have had authorization in substance. Password sharing, standing alone, is not a federal crime. The question is always whose permission the law requires, and whether it was given.
03Gates-up-or-down, and why it cuts both ways
Van Buren described the statute's structure with a now-famous metaphor: liability under the relevant provisions stems from a gates-up-or-down inquiry. Either you can access a computer or area of a computer, or you cannot. The appeal of that framing is its clarity. It replaces a vague inquiry into motive with a concrete one about whether a particular door was open to this user.
Defendants in access cases invoke the metaphor constantly, and often correctly: if the gate was up, motive does not matter. But the metaphor is not a safe harbor for anyone who can get a gate to open. It asks whether the gate was up for that user. A gate lifted because a credential belonging to someone else was presented is not obviously a gate that was up for the person presenting it. Van Buren's holding protects the entitled user who misuses what is behind the gate. It says much less about someone who was never issued a key.
This is why a technically successful login proves less than it appears to. Authentication is a statement about whether a secret matched a record. Authorization is a statement about whether the system owner consented to this person's access. The two usually coincide, which is why they are easy to conflate, and the gap between them is precisely where the hard CFAA cases live.
04The outsider problem: whose permission counts
Suppose, hypothetically, that a company wanted to see a competitor's ratings platform from the inside and could not get an account of its own. Suppose it identified an existing subscriber, arranged for that subscriber's credentials to be made available, and used them. The subscriber consented. The platform operator did not, and was not asked.
That is the configuration the courts have found hardest. One answer is that the account holder's consent is enough: the credential was lent, not stolen. The other is that the operator alone decides who may use its systems, and lending a key does not transfer the operator's consent to a stranger, particularly a competitor the operator would plainly have refused. Courts have leaned toward the second answer where the user is an outsider who knew the operator would not have permitted the access, and toward the first where the arrangement was ordinary, tolerated, or within the account holder's own authority.
In the Factor litigation, the complaint and the text messages filed as Exhibit D describe an arrangement in which credentials associated with an account under the name Emma Mason Furniture were allegedly used to reach a competitor's systems over a VPN. Those are allegations. Nothing published establishes who used the account, what the platform operator knew or permitted, what was opened, or whether any payment was made for access. The defendants deny the claims and have not been found liable for anything. As the analysis of federal discovery in this matter explains, these are exactly the questions logs, subpoenas and sworn testimony exist to answer.
05Which subsections are actually in play
“CFAA liability” is not one thing. The statute contains distinct offenses with different elements, and conflating them produces bad analysis.
§ 1030(a)(2)(C) — intentionally accessing a computer without authorization or exceeding authorized access, and thereby obtaining information from any protected computer. The broadest hook, and the one Van Buren construed. Ordinarily a misdemeanor, with felony treatment in defined circumstances including commercial advantage or private financial gain.
§ 1030(a)(4) — knowingly and with intent to defraud accessing a protected computer without authorization or exceeding authorized access, and by that conduct furthering the fraud and obtaining anything of value. It adds a fraudulent-intent element and a value threshold.
§ 1030(a)(6) — knowingly and with intent to defraud trafficking in passwords or similar information through which a computer may be accessed without authorization, where the trafficking affects interstate or foreign commerce or the computer is used by or for the United States. This is the provision that speaks to credentials changing hands, and it too requires intent to defraud.
§ 1030(g) — the civil action. A private plaintiff must suffer damage or loss and satisfy a qualifying factor, most commonly loss aggregating at least $5,000 in value during a one-year period. “Loss” means the cost of responding, assessing damage and restoring systems, plus revenue lost from any interruption of service. Reputational harm and competitive disadvantage are not loss. Recovery is compensatory; the statute provides no damages multiplier.
The civil threshold matters more than it sounds. A platform operator that never investigated, because it never knew, may have incurred no qualifying loss at all. Conversely, an operator that ran a serious incident response may clear $5,000 within days. Nothing in the public record establishes whether any such response occurred here.
06Nosal, Durenleau and Eddings
Three appellate decisions frame how courts are applying the authorization question after, and in one case before, Van Buren.
In United States v. Nosal (9th Cir. 2016), a former employee whose own access had been revoked worked with former colleagues who logged in using a current employee's credentials, which she had shared, to pull data from the employer's database. The Ninth Circuit affirmed his convictions, holding that once the employer rescinded authorization, a password borrowed from a willing insider did not restore it. Nosal is the closest pre-Van Buren statement that the system owner's permission, not the account holder's, is what the statute requires. It drew a vigorous dissent warning about criminalizing password sharing, and Van Buren did not pass on it.
In NRA Group, LLC v. Durenleau (3d Cir. 2025), the Third Circuit went the other way on very different facts. An employee out sick asked a colleague to retrieve a document from her work computer, which the colleague did with her express permission. Both had their employer's permission to use the computers, and the court held that violating workplace computer-use policies did not turn authorized use into CFAA liability. The opinion opens by observing that in the wrong hands the law becomes a hammer in search of a nail. Its rule is that an employee is authorized to access her employer's computer when the employer gives her permission to use it.
In United States v. Eddings (3d Cir. 2025), the same court addressed when authorization ends. A worker who had been given access to a board member's email account kept using it after she resigned, and Eddings was convicted as her accomplice and co-conspirator. Because the employer had taken no step to rescind permission, and no contract tied the authorization to her employment, the court vacated the conviction for aiding and conspiring in accessing a computer without authorization. The lesson is procedural as much as doctrinal: a system owner who wants access to end must actually end it, and prosecutors cannot infer revocation from a soured relationship.
Read together, these cases point in a consistent direction. The owner's permission is the touchstone. Courts will not manufacture a lack of authorization out of policy violations, bad motives, or relationships that went wrong. But where the owner never gave permission to the person who used the system, and would not have, the gates-up-or-down framing does not help that person. Nothing in Van Buren, Durenleau or Eddings immunizes an outsider who gets in on someone else's key.
07Trade secrets are a separate question
Access liability and trade-secret liability are different inquiries with different elements, and a case can succeed on one while failing on the other. Under 18 U.S.C. § 1839, information is a trade secret only if the owner took reasonable measures to keep it secret and it derives independent economic value from not being generally known or readily ascertainable. Misappropriation requires acquisition by improper means, or use or disclosure by someone who knew the information was improperly acquired.
Three consequences follow. First, lawful access does not defeat a trade-secret claim: a subscriber entitled to view data can still misappropriate it by using it in a forbidden way. Second, unlawful access does not establish one: if what was seen was not a protected secret, there is nothing to misappropriate. Third, criminal trade-secret exposure under the Economic Espionage Act is a separate track again, requiring proof beyond a reasonable doubt and a prosecutor willing to bring it. No charges or investigation are publicly known in this matter.
08What the evidence would have to show
Because the legal question turns on the operator's permission rather than on whether a login worked, a court would need a specific evidentiary record. The checklist below is what serious litigation over these allegations would be built on.
Evidence a CFAA authorization claim would require
Question
Evidence that answers it
Whose credential was used
Account registration records, billing records and the identity the operator believed it was dealing with.
What the operator permitted
Terms of service, account provisioning records, and any notice revoking or limiting access.
Who actually logged in
Authentication logs, source addresses, device fingerprints, session timing and VPN records.
What was opened or exported
Page-level access logs, API calls, report downloads and export volumes.
Whether entitlements were exceeded
Entitlement configuration for that account versus the areas actually reached.
Knowledge and direction
Communications about obtaining or paying for access, and who approved it.
Loss
Documented investigation, remediation and response costs within a one-year period.
Two cautions apply to all of it. Civil and criminal exposure are distinct: a civil plaintiff must prove its case by a preponderance of the evidence and must show qualifying loss, while a prosecutor must prove every element beyond a reasonable doubt and must choose to charge. And the allegations in this matter are disputed. SecurityScorecard and Aleksandr Yampolskiy deny them, no court has made any finding, and this analysis takes no position on what happened.
Scenario B and the hypothetical in section four are illustrations of a legal category, not assertions about any person or company. Allegations described here are unproven, the defendants contest them, and nothing in this article should be read as a conclusion about anyone's liability or guilt.
The FBI’s Boston Cyber Task Force interviewed multiple witnesses about Yampolskiy’s alleged access to Bitsight and requested evidence tracing Bitsight material into SecurityScorecard’s product.
A day after his firm sent preservation letters to a Factor executive and an advisor, Yampolskiy’s defense attorney Michael Yaeger demanded that the Chronicle come down. His email identified no factual error in the articles he attached.
On November 19, Carlton Fields attorney Michael D. Margulies emailed a Factor Cybersecurity executive and an advisor at their personal addresses. They were already represented by counsel.