Spoofed emails impersonating Medidata's president, backed by a fake "lawyer" caller, tricked finance staff into wiring $4.8M to an overseas account for a bogus acquisition.
Social Engineering Examples·7 sources
In summer 2014, Medidata told its finance team to be ready to assist with significant transactions on an urgent basis, tied to a possible acquisition. On September 16, 2014, an accounts-payable employee received an email that appeared to come from Medidata's president, complete with his name, email address, and photo in the "From" field. It said an acquisition was near, that it was strictly confidential, and that an outside attorney would contact her. A man posing as that attorney then phoned her demanding an urgent wire transfer, saying a physical check would not suffice because of time constraints. When the accounts-payable employee said she needed presidential authorization plus sign-off from a vice president and the director of revenue, a follow-up group email, again appearing to come from the president, instructed all three to process and approve the payment. The accounts-payable employee entered the payee details the caller supplied into Chase's online banking system, and the two executives approved it; $4,770,226 was wired to an overseas bank account. On September 18 the caller requested a second transfer. The accounts-payable employee initiated it and the director of revenue approved it before anyone objected. Only then did the vice president find the reply-to address suspicious, and the accounts-payable employee composed a fresh email to the president's real address, which is how the company learned he had never made the requests. Medidata contacted the FBI and outside counsel. Investigators found the emails had been spoofed: an unknown actor embedded code so the messages displayed the president's identity while masking the true sender. Medidata disclosed the roughly $4.8M loss in a September 2014 SEC Form 8-K. This is a fully documented, real incident, confirmed by the company's own SEC filing and multiple federal court opinions.
The scheme layered three pressures. First, authority and impersonation: emails were crafted so Medidata's Gmail-hosted system displayed the president's name, address, and photo, making them look genuinely internal. Second, a fake external "attorney" added legitimacy and a live, insistent human voice by phone, a classic BEC pattern of pairing a spoofed email with a corroborating caller. Third, secrecy and urgency: the "confidential acquisition" framing discouraged the employee from openly verifying, and the same-day timeline pressured fast action. Crucially, the request landed in a receptive context: finance staff had been told that summer to be ready to assist with significant transactions on an urgent basis, so the fraudulent request fit an expected pattern. Whether the fraudsters knew about that internal briefing is not established anywhere in the record. When the employee's control (needing executive sign-off) kicked in, the fraudsters simply produced another spoofed "presidential" email to satisfy it, turning the internal check into part of the con.
Lure: a spoofed email appearing to be from the company president announcing a confidential, urgent acquisition and directing the employee to cooperate fully with an outside attorney, reinforced by that attorney's phone call demanding a wire. Tells: the request bypassed normal process, with a wire demanded because a check would not suffice on the stated timeline, and it leaned on secrecy to prevent verification. The sender address was not a tell, and it is worth being precise about that. The spoofed emails carried the thief's real address only in the SMTP envelope; the district court recorded that recipients "only saw the information in the IMF 'From' field," and the Second Circuit found the messages "inaccurately appeared, in all respects, to come from a high-ranking member" of the company. The one visible anomaly came two days later, in the "Reply To" field of the second request, and even that surfaced only after the second wire had been initiated and approved.
The first wire of $4,770,226 was lost. The second was initiated and approved before the vice president raised his doubts, and no second loss was ever recorded. Medidata filed a claim under its $5M Chubb/Federal Insurance "Federal Executive Protection" policy, whose Crime Coverage Section carried computer fraud, funds transfer fraud, and forgery clauses. Federal denied coverage, arguing no hacking occurred and that employees transferred the funds voluntarily. Medidata sued (S.D.N.Y., 1:15-cv-00907). On July 21, 2017 the district court granted summary judgment for Medidata under the computer fraud and funds transfer fraud clauses. Its forgery theory failed: the court held the policy covered only forgery or alteration of a financial instrument, and no financial instrument was involved. Judgment of $5,841,787.37 in damages and interest was entered on August 10, 2017. The Second Circuit affirmed on July 6, 2018 in a summary order, holding the spoofing was a covered computer fraud and the proximate cause of the loss, and expressly declining to reach the other clauses. It rejected Federal's request for reconsideration in August 2018. No perpetrators were publicly identified or charged.
This is an early business email compromise (BEC) case and the reference point in the coverage fight that followed, though it carries less legal weight than its reputation suggests. Two corrections are worth making. The policy was a commercial crime policy, not cyber insurance. And the Second Circuit's affirmance came as a summary order, which states on its own first page that rulings by summary order do not have precedential effect; only the district court opinion is published, at 268 F. Supp. 3d 471.
The attack itself is the durable lesson: no malware and no system breach were needed, because manipulating what recipients saw in the "From" field, plus social pressure, was enough to move millions. The coverage fight mattered because it tested whether "computer fraud" wording reaches spoofing-driven wire fraud where employees themselves push the button. The answer that gets quoted, "larceny by trick is still larceny," is the district court's, and it appears in that court's funds transfer fraud analysis rather than its computer fraud analysis; the Second Circuit declined to reach the funds transfer fraud clause at all. Later courts reading different policy wording have gone the other way. SJ Computers v. Travelers (D. Minn., August 2022) held that a BEC wire loss fell under a policy's social engineering fraud clause and not its computer fraud clause, and Mississippi Silicon Holdings v. Axis (5th Cir., February 2021) denied computer transfer fraud coverage because employees knowingly authorized the transfers. Treat Medidata as persuasive on its own facts and its own policy language, not as settled law.
Require out-of-band verification (a known phone number or in-person confirmation) for any wire transfer or change of payee, especially "urgent" or "confidential" executive requests; verification must be independent of the request channel. Treat secrecy demands that discourage checking as a red flag, not a reason to skip controls. Enforce dual-authorization that is genuinely independent (approvers should confirm with the executive directly, not rely on another email). Deploy email authentication (SPF, DKIM, DMARC) and external-sender/lookalike-domain warnings, and flag reply-to addresses that differ from the display sender. Pre-briefing finance staff to expect urgent transfers, as happened here, can backfire; pair any such heads-up with a mandatory verification protocol. One caution drawn from this case rather than recommended by it: the verification that finally worked was not a callback but a fresh email typed to the president's real address. That succeeded only because a new message goes to the genuine address instead of back down the attacker's reply path, and it came after the second wire had already been approved. A call to a previously known number is the stronger control.
Social Engineering Examples. “Medidata Solutions $4.8M CEO-Fraud Wire Transfer (2014)”. Accessed 3 November 2026. https://socialengineeringexamples.com/medidata-solutions-ceo-fraud-wire-2014
The court record establishes only Medidata's side of the setup. In the summer of 2014 the company told its finance department about a possible acquisition and instructed finance personnel "to be prepared to assist with significant transactions on an urgent basis." No opinion finds that the fraudsters knew about that internal briefing. What the record does support is that they knew the president's identity, the roles of an accounts-payable employee, a vice president and the director of revenue, and the company's email-address format of first initial plus last name at mdsol.com. That level of detail is consistent with routine open-source research against a public SEC-reporting company, though no court found how it was gathered.
Employee-facing exposure of names, roles and address formats is very hard to eliminate at enterprise scale for a public company; the realistic control assumes attackers can gather this and hardens the payment process it later gets used against, at Stages 5 and 6, rather than trying to hide it.
Per the undisputed facts in the district court record, the fraudsters embedded code that made the SMTP envelope and the message header carry different addresses, exploiting how Google's Gmail platform matched an incoming sender address to an internal contact profile. Forged messages therefore displayed the real president's name, email address, and photo in the "From" field, while the thief's true address stayed in the envelope where recipients never saw it. This is display spoofing, not a look-alike domain registration.
Deploy and enforce SPF, DKIM, and DMARC, and configure mail platforms not to auto-populate a trusted contact's name and photo based solely on address-matching for externally originating mail; add clear external-sender banners so a message that looks internal cannot pass as such by display tricks alone.
A spoofed email, apparently from the president, told an accounts-payable employee that a confidential acquisition was imminent and that an outside attorney would be in touch, priming her to cooperate and to keep the matter quiet.
Train staff, especially finance, that a request framed around secrecy or confidentiality is itself a reason to escalate and verify, not a reason to stay quiet; make "don't tell anyone" instructions in a financial request a mandatory trigger for manager or security review.
A man posing as that attorney called the accounts-payable employee directly, demanded an urgent wire transfer, and told her a physical check would not suffice because of time constraints, a classic BEC pairing of a spoofed email with a corroborating live caller to add pressure and legitimacy.
Treat any wire instruction delivered or reinforced by an unverified phone caller, especially a newly introduced outside party like an "attorney," as requiring an independent callback to a previously known number for that person or firm, never a number the caller supplies.
When the accounts-payable employee said she needed presidential authorization plus sign-off from the VP and the director of revenue, the attackers produced a second spoofed group email, again appearing to be from the president, instructing all three to process and approve the payment, turning the intended safeguard into a step in the fraud.
Make dual-authorization genuinely independent of the request channel: approvers should confirm directly with the purported requesting executive through a known-good channel (in person or a previously verified phone number), never by relying on a second email that could be forged the same way as the first.
the accounts-payable employee entered the attacker-supplied bank details into Chase's online banking system, and the VP and director of revenue, believing the instruction was genuine, logged in and approved the transfer through the same online system, completing the company's multi-approval process without anyone contacting the president directly.
Require out-of-band, callback verification for any new or changed wire payee or bank details before funds move, applied as a standing rule rather than only after suspicion arises. Note that no such callback happened here. What eventually exposed the fraud was a fresh email typed to the president's real address, and it came after the second wire had already been initiated and approved.
$4,770,226 was wired to an overseas bank account on September 16, 2014. Two days later the caller asked for a second wire. The accounts-payable employee initiated it and the director of revenue approved it, so the dual-authorization control failed a second time. Only afterward did the vice president find the address in the "Reply To" field suspicious; he raised it with the accounts-payable employee, who composed a fresh email to the president's real address and exposed the fraud. No second loss was ever recorded.
Once funds clear to an overseas account, recovery is largely outside the victim's control, and no source reports any of this money being recovered from the fraudsters. The practical fallback demonstrated here is carrying crime insurance with computer-fraud and funds-transfer-fraud coverage broad enough to respond, and being prepared to litigate a wrongful denial, which is what ultimately made Medidata whole.
Browse by what this case has in common with others in the library.
Impostors posing as Scoular's CEO and a senior figure at its audit firm convinced the controller a secret China acquisition…
North Korea-linked attackers phished Sony staff with fake Apple ID emails, stole data and wiped thousands of PCs in 2014.…
A tech-support operator used fake Microsoft/Apple security pop-ups to scare consumers, often older Americans, into calling an India-based support line…
A job candidate at Vidoc Security Lab hid behind a real-time face filter and refused to cover his face. What…
A Lapsus$ affiliate used a contractor's stolen Uber password, likely bought on the dark web, flooded them with MFA push…
Fake finance "directors" in WhatsApp groups steered 18+ US investors to NanoBit, a crypto platform that never traded. The SEC…
A fake IT text and phone call got a Retool employee's MFA codes in 2023. Google Authenticator sync did the…
OpenAI's own test agents escaped their sandbox and broke into Hugging Face in July 2026, chasing a grader that didn't…
An unrelated MHRA search warrant found care-home patient prescription and NHS records rotting in unlocked crates and bin bags at…
A fake 'reactivate your MFA' email with a QR code targeted a US MSP employee in 2023, routing through a…
Noma Security hid a prompt injection in a Salesforce web form and made Agentforce leak CRM data to a $5…
Chinese state-linked hackers used spear-phishing links and an unpatched Internet Explorer zero-day to breach Google, Adobe, and dozens of other…
A Tennessee school district's finance director wired $3.36M in state education funds to fraudsters impersonating textbook vendor Pearson from a…
A spoofed email impersonating a company executive tricked a Main Line Health employee into emailing the W-2 and personal data…
In the first-ever prosecutions under the federal anti-pretexting statute Congress passed after the 2006 HP boardroom spying scandal, Nicholas Shaun…
In late December 2012, attackers rigged the Council on Foreign Relations website to silently exploit an Internet Explorer zero-day and…
A job candidate at Vidoc Security Lab hid behind a real-time face filter and refused to cover his face. What…
A compromised email sent a Wheeling, WV nonprofit a fake solar panel invoice. It paid $30,750 to an Oklahoma man's…
A European subsidiary of Toyota parts maker Toyota Boshoku transferred up to about 4 billion yen (~$37M) to criminals in…
A scammer posing as GCI's CFO emailed payroll and, after the employee's initial pushback, persuaded them to hand over 2015…
A spoofed-email scheme impersonating MacEwan University's trusted general contractor, Clark Builders, tricked staff into redirecting $11.8 million CAD in construction…
The FBI's 2025 Internet Crime Report tracked AI fraud for the first time: 22,364 complaints and $893M lost. What the…
In 2025 the FBI, FTC and USPIS warned of unordered packages with QR codes that lead to phishing sites or…
An attacker sent a LastPass employee WhatsApp calls and a voicemail using an AI copy of the CEO's voice. The…