Encypher

🚨 34 newspaper publishers. Nearly 400 local outlets. One lawsuit. 🚨 A coalition of local and regional newspapers just sued over how their journalism was scraped to build and power AI products. Buried in the complaint is the detail that matters most: the allegation that copyright management information was STRIPPED OUT! >Author credits. >Publication names. >Copyright notices. The exact signals that say who made a story and who owns it. That is the real wound. Not only that the work was copied, but that its origin was ERASED. Here is the uncomfortable part for every publisher watching this case. A robots.txt file is a sign, not a lock. It sits at your front door and does nothing once a story is copied somewhere else. We audited 457 news sites. More than half earned an F. Four out of five let the answer-writing bot walk right in. 🚩 👉 And one of the publishers in this very lawsuit BLOCKS ALL 13 known AI crawlers, CORRECTLY, and still says its work was taken. Provenance is the other path. You put a cryptographic mark on the story itself, not just a rule on your server. It carries proof of who published the work and when, and it travels with the content across the open web. That gives you verifiable evidence of ownership, instead of a request that stops at your own front door. We could have helped these publishers put that evidence in place before the lawyers got involved. We can still help the next 400. Litigation is the failure mode. Provenance is the prevention. Free 60-second check on your own site: https://lnkd.in/eb3jiA4w Encypher is content authenticity infrastructure built on C2PA, the open industry standard for proving where content comes from. We co-chair the C2PA Text Provenance Task Force. Our job is to give publishers verifiable proof of what their newsrooms created, so ownership is something you can demonstrate, not just claim. If you want to learn more, DM's are open.

Every week, friends and family send me viral clips. A politician saying something outrageous, a cool ad, or just a funny video. The question is always the same: how do I know this is real and not AI? Who do you trust? That is the entire reason we exist. And today, Encypher is part of the answer. This week, we passed C2PA conformance and were added to the official C2PA Trust List as both a Certificate Authority and a Time Stamp Authority. In plain terms: content signed with an Encypher certificate now verifies as trusted in any C2PA-compliant tool, anywhere it travels. We can issue trusted signing certificates to publishers and enterprises, and trusted timestamps, from our own root. C2PA is the standard behind content authenticity at Adobe, Google, Sony, OpenAI, BBC, Microsoft, The New York Times, and hundred of others. You do not apply for a spot on that trust list and receive it. You earn it through conformance, audits, and cryptographic rigor that leaves no room for "close enough." As Co-Chair of the C2PA Text Provenance Task Force, we helped write the text spec. Now we operate full-stack trusted infrastructure for it. The proof of origin our partners embed today is recognized as trusted everywhere it goes tomorrow, for almost any digital asset. We started Encypher to give content a verifiable origin in a world drowning in synthetic media. Becoming a root of trust for that origin is the most meaningful milestone we have hit. To our team, our advisors, and the Coalition for Content Provenance and Authenticity (C2PA) community: thank you. This was a huge milestone. If you sign or verify content and want to understand what trusted C2PA infrastructure unlocks, let's talk. #C2PA #ContentAuthenticity #ContentProvenance #AITransparency #TrustList󠇟󠇠󠇡󠇢󠅕󠇠󠄏󠇄󠄧󠄔󠄸󠇇󠅃󠄏󠇎󠆨󠄧󠆌󠅂󠄔󠅈󠇢󠅁󠆄󠇅󠅳󠅔︂󠆎󠄶󠅓󠆨󠆩󠇕󠄵󠆩󠆷󠆻󠆷︅󠄵󠆚󠆹󠄡

The part of David Evan Harris's post that sticks with me: the tools to mark AI text have existed for a while. The reason students, teachers, and readers couldn't use them was a choice, not a technical limit. We built Encypher to make the better choice the easy one. When text is generated, we embed an invisible, cryptographically signed mark directly in the characters. It carries the signed provenance itself, not just a pointer to it, and anyone can check it with an open verifier. That is both the signed-metadata layer and the publicly readable "signpost" the EU's new Code of Practice describes, working today, in plain text. On 2 August, marking AI-generated text in the EU stops being optional. The technology is ready. The standard is here. What's left is the decision to use it. Grateful to David for seeing this early, and for pushing it where it counts. #ContentProvenance #AITransparency #EUAIAct #C2PA #ResponsibleAI󠇟󠇠󠇡󠇢󠄫󠅪󠆅󠄇󠄬󠅕󠆲󠄮󠅸󠄃󠅵󠇍󠅲󠅠󠇊󠆁󠅸󠆜󠅭󠇝󠆔󠅱󠅇󠆅󠅿󠆞󠇠󠄺󠇒󠅜󠅰󠄍󠅯󠄤󠄗󠅺󠆎󠄀󠅪󠇅

𝗪𝗲 𝗽𝗮𝘀𝘀𝗲𝗱 𝘁𝗵𝗲 𝗖𝟮𝗣𝗔 𝗖𝗼𝗻𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝗣𝗿𝗼𝗴𝗿𝗮𝗺 𝗳𝗼𝗿 𝘁𝘄𝗲𝗻𝘁𝘆 𝗳𝗶𝗹𝗲 𝗳𝗼𝗿𝗺𝗮𝘁𝘀. Independently tested and passed under the C2PA Conformance Program. The most of any single signer product to date. Encypher is proud to now provide conformant singing + verification across 11 image, 3 video, 5 audio, and 1 document formats. The generator security review and the implementation review both passed. In addition, our signing core signs 44 total formats to the current C2PA spec, the 20 the program tests, plus 24 more formats it doesn't cover yet. So we're doing two things at once: shipping what an C2PA has verified, and helping the program build conformance tests for the modalities it doesn't cover yet including live video and the text spec I co-authored into the standard. Conformance turns "we sign content" into "an independent body confirmed we do it correctly, and we're interoperable with the ecyosystem." That's the point of provenance infrastructure: the proof has to be checkable by someone other than the vendor making the claim. Text conformance doesn't exist yet. We're building it for Coalition for Content Provenance and Authenticity (C2PA) and the Content Authenticity Initiative. We're proud partners. P.S. Thank you to Scott Perry and Sherif Hanna for the reviews and the rigor. The program is demanding on purpose, and the ecosystem is stronger for it. #C2PA #ContentProvenance #ContentAuthenticity #TextProvenance󠇟󠇠󠇡󠇢󠇔󠄅󠇇󠇯󠆷󠇢󠇗󠇆󠄽󠅎󠅗󠇙󠅁󠅉󠇂󠇗︌󠅁󠆎󠅺󠄅󠇝󠄢󠅒󠆯󠄑󠄜󠆎󠄱󠄊󠄯󠅴︅󠄘󠆍󠄣󠅑󠅈󠇇󠄭

Appreciate the kind words from Trust In Media Cooperative. Their closing question gets at something we encounter every day in the C2PA standards work: the credibility signals that will matter most are the ones that cannot be separated from the content. Labels can be removed. Most metadata gets stripped during distribution. Terms of service go unread. Every intermediary between origin and destination is a point where context disappears. The C2PA text standard published in January takes a different approach. Cryptographic proof of origin embedded directly in the text. It survives copy-paste, redistribution, and scraping. The provenance travels with the content wherever it goes, regardless of who handles it. That is the infrastructure shift. Not better labels. Provenance that persists. #C2PA #ContentProvenance #AI #Provenance

The IP question in film is the same question we've been working on in publishing and news: when content can be copied, summarized, scraped, and redistributed at machine speed, how does the creator prove it was theirs? And, how does the EU AI act impact film? Carlo and the CinemaO team assembled 28 contributors across legal, production, technology, and investment to work through this. Proud that Encypher is both a supporter of the publication and a featured case study, the piece we contributed is itself a working demonstration of the technology. The content in the white paper carries cryptographic provenance signatures that anyone can verify. Looking forward to the panel on May 18th at 11:30am at the Marche du Film main stage with Angela Dunning, Bastian Bergmann, Elisa Alvares, and Yannick Bossenmeyer. If you're at Cannes, come find me, we'll also be at the Opening Night at Village Innovation tonight and again on May 17th, 10am!

The EU AI Act comes into force August 2. Most publishers assume it only applies to OpenAI and Google. It doesn't. If your newsroom uses AI for headlines, summarization, translation, or content recommendations, you qualify as a deployer under the Act. Disclosure obligations apply to you too. The Hamburg court ruling in December made the direction clear: robots.txt is not technical infrastructure. Regulators and courts expect machine-readable proof. Three things publishers should do before August 2: 1. Audit AI tool usage across editorial and distribution workflows 2. Implement C2PA content provenance - free to sign, free to verify 3. Serve formal notice to AI companies that your content carries cryptographic signatures The same infrastructure that satisfies EU compliance strengthens your licensing position. California's parallel requirements take effect the same day. Full breakdown on our blog: Link in comments #EUAIAct #C2PA #ContentProvenance #PublishingIndustry #AIRegulation