404 Media

404 Media is a journalist-owned investigative technology outlet founded on August 22, 2023, by Jason Koebler, Samantha Cole, Emanuel Maiberg, and Joseph Cox — all previously senior editors and writers at Vice Media's Motherboard division, which they left after Vice's bankruptcy filing. The outlet is structured as a reader-subscription cooperative modeled on Defector Media and Hell Gate, explicitly rejecting venture capital in favor of a reader-supported model with paid subscribers receiving full article access, ad-free podcasts, and FOIA brainstorming sessions. It became profitable by February 2024.

The outlet's AI-harm beat is one of the most consequential in independent technology journalism. Co-founder Samantha Cole first reported on deepfake non-consensual intimate imagery in December 2017 at Motherboard — the first journalist to cover the phenomenon — and her reporting was credited with prompting Twitter and Reddit to update their platform rules against non-consensual synthetic pornography. That investigative lineage continued at 404 Media: within its first year the outlet's deepfake reporting directly triggered Apple to remove three NCII-enabling applications from its App Store after the investigation provided links to the apps and their associated advertisements, and a bipartisan group of members of Congress sent letters in December 2024 to Apple CEO Tim Cook citing the 404 Media investigation and questioning why Apple's own review process failed to catch the applications. A separate 404 Media investigation into non-consensual AI-generated intimate images of Taylor Swift — traced to distribution networks on 4chan and Telegram — prompted a congressional letter to Microsoft CEO Satya Nadella.

Founding and model

The four founders departed Vice in the context of that company's May 2023 bankruptcy filing, launching 404 Media on August 22, 2023, with what Fast Company described as "a spartan setup consisting of a Stripe account and the Ghost web-hosting platform." The model was deliberately anti-venture-capital: the founding announcement articulated a commitment to paying journalists livable wages without growth-at-any-cost investor pressure, with subscriptions priced at $10/month or $100/year. Articles relying primarily on public records remain free to readers. The outlet became profitable within six months of launch.

The four founders brought complementary investigative specialisms: Koebler as a technology policy and right-to-repair reporter and editor-in-chief; Cole as the field's foundational deepfake and online-harm journalist; Maiberg as a senior editor covering AI and online culture; and Cox as an award-winning hacking and cybercrime investigative reporter whose prior work had triggered hundreds of millions of dollars in fines against telecoms and shut down data brokers. In November 2024 the outlet entered a co-publication arrangement with Wired, co-publishing two articles monthly.

AI harm investigations

404 Media's AI beat covers the intersection of generative AI, synthetic media, platform accountability, and harms to real people — particularly non-consensual intimate imagery, deepfake-enabled fraud, and AI-generated misinformation.

Non-consensual intimate imagery (NCII). The outlet's NCII reporting is anchored in Samantha Cole's decade-long beat. In 2023, 404 Media conducted a collaborative investigation with UC Berkeley digital forensics expert Hany Farid into a deepfake pornography platform. The AI undress app series, which documented how easily available applications on Google Play and Apple's App Store enabled the creation of non-consensual nude images, directly produced the Apple removal outcome above and prompted the congressional response. The EFF's 2024 award citation specifically named the outlet's reporting on "non-consensual AI-generated pornography platforms" as among the work warranting recognition.

Deepfake fraud infrastructure. In May 2026, Joseph Cox obtained a copy of Haotian AI, a Chinese-origin real-time face-swap application marketed to scammers. Cox tested the software on a live Teams call, confirming it could sustain convincing real-time impersonation across WhatsApp, Zoom, and Microsoft Teams — the infrastructure behind a new class of video-call fraud. The investigation documented a commercially available, actively deployed tool enabling synthetic identity fraud at scale.

Synthetic content and platform accountability. A January 2024 investigation documented AI-rewritten versions of 404 Media's own articles appearing on spam sites and ranking above originals in Google Search. A February 2024 report alleged Tumblr and WordPress had sold user data to OpenAI and Midjourney for model training without clear user consent. A separate investigation into Amazon's sale of AI-generated books dangerous to consumers prompted Amazon to remove the titles.

Surveillance and biometrics. In June–July 2025, the outlet reported on ICE's Mobile Fortify application, which used a 200-million-image database for contactless biometric identification in the field — part of a broader beat on law-enforcement AI deployment that draws on FOIA requests as a primary evidence tool.

Recognition

In 2024 the Electronic Frontier Foundation presented 404 Media its Pioneer Award for Fearless Journalism, describing the outlet as having "tirelessly forged ahead with incisive investigative reports, deep-dive features, blogs, and scoops" on hacking, cybersecurity, cybercrime, AI, consumer rights, surveillance, and privacy. The Financial Times highlighted 404 Media as a successful independent news operation amid broader media-industry contraction. The outlet's reporting on AI-generated NCII has appeared in academic scholarship (Sage Journals 2025 paper on AI undressing apps) as a documentary source for the normalization-of-harm phenomenon the investigations surfaced.