Former Mayo Clinic Director Alleges AI Research Misconduct, Data Manipulation, and Retaliation in Federal Lawsuit
核心洞察
Traci Tamiko Eto, former Mayo Clinic Director of Research Operations, filed a federal lawsuit alleging she was fired after flagging serious AI safety violations including data manipulation and patient privacy breaches.
The lawsuit claims researchers on Mayo's MAYA (搜索) digital assistant concealed a 67% error rate by deleting unfavorable results, with 10 internal whistleblower reports raising similar alarms.
Eto alleges she uncovered unauthorized AI deployment in clinical workflows, unapproved cardiac surgery using an investigational device abroad, and improper genome data sharing without patient consent.
A former Mayo Clinic director has filed a federal lawsuit against the Rochester-based health system, alleging she was terminated after raising concerns about serious lapses in the institution's artificial intelligence research practices, including data manipulation, patient privacy violations, and the unauthorized use of AI tools on patients.
Traci Tamiko Eto filed the complaint on July 6 in U.S. District Court in Minnesota, accusing Mayo Clinic of retaliation, disability discrimination, and violations of the Family and Medical Leave Act (FMLA). Mayo Clinic has not yet filed a formal response to the lawsuit, though the health system said in a written statement that it is "committed to the responsible development and deployment of AI, with privacy, security, transparency and compliance embedded throughout our processes."
Allegations of Systematic Compliance Failures
Eto joined Mayo Clinic in December 2023 as Director of Research Operations, overseeing a team of 36 employees and three managers. According to the complaint, she was brought on specifically to help the clinic align with a 2023 White House executive order establishing AI governance standards for federal contractors.
Over the subsequent 18 months, the lawsuit alleges Eto uncovered a series of compliance failures within Mayo's AI and research programs. Central among these is the claim that researchers working on Mayo's AI digital assistant, called MAYA (搜索), deleted unfavorable results and misrepresented outcomes in an attempt to conceal a 67% error rate. The lawsuit states that 10 separate internal whistleblower reports raised similar alarms.
The complaint further alleges that de-identification processes used when sharing patient data with global partners were bypassed without proper review by Mayo's Institutional Review Board (IRB), a federally required oversight body. "The IRB process is absolutely critical to research integrity, and it's also absolutely critical to patient privacy and to patient safety," said Artur Davis, an employment law attorney representing Eto. "And if that process is not working the way it's supposed to, you literally have garbage in, garbage out, potentially."
Additional allegations include the injection of an unsanctioned software tool into ordinary clinical workflow without regulatory authorization, compromising both patient safety and data security. The suit also claims a supervisor authorized a high-risk investigational medical device to perform cardiac surgery in a foreign country, even though the procedure had never received IRB approval and was not licensed for use there. Eto's team also allegedly uncovered a plan to distribute patients' genome sequencing data into a widely accessible database, despite the original patient consent forms not authorizing that kind of sharing.
Retaliation and Termination
According to the complaint, Eto's standing at Mayo deteriorated after she pushed back on these practices and filed an internal report with the clinic's legal department in February 2025. The lawsuit alleges that within days of supervisors being briefed on her report, Eto was excluded from executive meetings and replaced by a subordinate. An AI engineering director reportedly warned her that a senior supervisor had issued "marching orders to get rid of Tamiko ASAP."
In March 2025, Eto received a $7,000 raise, bringing her salary to approximately $166,000. Weeks later, she was placed on a formal performance improvement plan. The stated reason, according to the complaint: she was a poor "cultural fit." The lawsuit says no specific performance failures were cited.
By July 2025, Eto had been stripped of her management responsibilities and removed from projects she had created, including an AI Protocol Builder she invented. She was also removed from the patent filing for that tool, the suit alleges, which diminished her financial stake in the invention.
The complaint states the professional pressure took a toll on Eto's health. She was diagnosed with symptoms of distress and chronic anxiety and applied for medical leave under the FMLA. The lawsuit alleges Mayo initially rejected her leave application, and that confidential details about her mental health were shared with management officials who had no role in the leave process. Mayo ultimately approved her leave after Eto retained an attorney who sent the clinic a legal threat letter.
On September 2, 2025, roughly halfway through her approved leave, Eto was notified that her position had been eliminated in what Mayo called a reduction in force. The lawsuit says she was the only employee whose job was cut. She applied for 15 internal positions during a 90-day window, receiving one interview. She was officially terminated December 1, 2025.
Legal Claims and Broader Implications
The lawsuit brings four counts against Mayo Clinic: retaliation under the False Claims Act, disability discrimination under the Americans with Disabilities Act, retaliation under the FMLA, and interference with FMLA rights. Eto's attorneys argue that because Mayo receives hundreds of millions in federal research funding — approximately $498 million in federal and state government research funding in 2025, according to the complaint — her efforts to enforce compliance with federal standards constitute protected activity.
Eto is seeking back pay, front pay, compensatory damages for emotional distress, punitive damages, and attorneys' fees. The complaint also asks the court to declare Mayo's actions unlawful and order the clinic to stop discriminatory employment practices. A jury trial has been demanded.
The lawsuit also alleges Mayo maintains a "ghost file" system that flags former employees who raised compliance concerns as "Not Eligible for Rehire" in external background check systems without the employee's knowledge.
The case sets the stage for a public legal confrontation between Mayo Clinic, Minnesota's largest employer, and one of its former research leaders, amid an explosion of AI usage in health care that is helping doctors with everything from notetaking to diagnostic evaluations — but also raising broader concerns over the accuracy of AI tools and the privacy of patient data.