US public companies have to describe their business and revenue in their SEC filings. These are the ones whose filings say they license data or content to AI developers, quoted word for word and linked to the filing.
Updated October 22, 2026. A filing that mentions licensing shows the company has a deal or a revenue line, not what a smaller company would be paid. Free to reuse under CC BY 4.0.
| Company | What it licenses | Amount in the filing | Buyer named | Filing |
|---|---|---|---|---|
| Adaptive Biotechnologies Corporation ADPT | T cell receptor (TCR) binding datasets and machine learning models | None stated | Harell Data Corp. | 10-Q, 2026-08-04 |
| AI Era Corp. AERA | short-form drama video library | $955,269 | Anyone Pictures Ltd | 10-Q, 2026-07-20 |
| Chegg, Inc. CHGG | learning content | None stated | None | 10-K, 2026-03-09 |
| CuriosityStream Inc. CURI | video and audio content library, plus partner content under revenue share | None stated | None | 10-K, 2026-03-12 |
| Gannett Co., Inc. (now USA TODAY Co., Inc.) TDAY | news content | None stated | Microsoft | 8-K, 2025-10-30 |
| Getty Images Holdings, Inc. GETY | visual content licensed under multi-year AI licensing deals | $40 million | None | 8-K, 2026-10-13 |
| Getty Images Holdings, Inc. GETY | visual assets (images, video) and associated metadata | None stated | None | 10-K, 2026-03-16 |
| GoPro, Inc. GPRO | opted-in subscribers' cloud-stored video content | None stated | None | 10-Q, 2026-08-10 |
| GoPro, Inc. GPRO | opted-in subscribers' cloud-stored video content | $2 million | None | 8-K, 2026-08-10 |
| John Wiley & Sons, Inc. WLY | research and learning content (books, journals), including some licensed from other publishers | $49.1 million | None | 10-K, 2026-06-24 |
| Morningstar, Inc. MORN | investment data and research | None stated | None | 8-K, 2026-08-25 |
| News Corporation NWSA | news and information content (Dow Jones) | None stated | None | 10-K, 2026-08-07 |
| People Inc. (formerly IAC Inc.) PPLI | publisher content (People Inc. brands) | None stated | None | 10-K, 2026-02-20 |
| Reddit, Inc. RDDT | Reddit data (user posts and comments) via data API and data transfers | $203 million | None | S-1, 2024-02-22 |
| Serve Robotics Inc. SERV | video and LiDAR data collected by delivery robots | None stated | None | 10-K, 2026-03-12 |
| Shutterstock, Inc. SSTK | metadata for images, footage, music tracks and 3D models (data offering) | None stated | None | 10-K, 2026-02-17 |
| Similarweb Ltd. SMWB | digital (web traffic and behavior) data | None stated | None | 6-K, 2025-08-12 |
| Tempus AI, Inc. TEM | de-identified multi-modal oncology data | $200 million | Pathos, AstraZeneca | 10-K, 2026-02-24 |
| Thomson Reuters Corporation TRI | Reuters news agency content | None stated | None | 40-F, 2026-03-05 |
| Yelp Inc. YELP | local business reviews and content via the Yelp AI API | None stated | None | 10-K, 2026-02-27 |
| Zedge, Inc. ZDGE | rights-cleared image, video and audio datasets from its creator communities (GuruShots, Zedge Marketplace) plus crowdsourced content | None stated | None | 10-Q, 2026-06-12 |
Pursuant to the Platform Agreement, we have agreed to license certain TCR binding datasets and machine learning models for training and use by Harell customers in consideration for specified percentages of the aggregate customer revenue from training, inference, and gross model inference activities, as set forth in the Platform Agreement.
Read the filing on sec.gov, read 2026-10-22
The Company generates revenue from licensing its short-form drama content library for artificial intelligence training purposes.
$955,269, AI training pilot licensing fees received from Anyone Pictures Ltd, nine months ended May 31, 2026: The Company received total short-form drama for AI training pilot licensing fees of $955,269 and $0 from Anyone Pictures Ltd during the nine months ended May 31, 2026 and 2025.
Read the filing on sec.gov, read 2026-10-22
In addition, we have licensed certain of our content to other entities who may use this content to train AI models that may compete with our products in the market.
The filing reports content licensing revenue of $10.8 million without saying how much of it is for AI.
Read the filing on sec.gov, read 2026-10-22
This includes our focus on data licensing for AI model training, which represents the vast majority of our AI content licensing activity.
Read the filing on sec.gov, read 2026-10-22
We are also very excited to announce this morning our newest AI licensing agreement, partnering with Microsoft on their upcoming launch of its Publisher Content Marketplace.
For use in AI products; the filing does not say the content is used for training.
Read the filing on sec.gov, read 2026-10-22
2026 growth distorted by ASC 606 timing effects tied to large multi-year AI licensing deals signed in 2025, with ~$40MM of revenue recognized on an accelerated basis in Q4 2025.
$40 million, approximate revenue recognized on an accelerated basis in Q4 2025 from large multi-year AI licensing deals signed in 2025 (filing says approximately): with ~$40MM of revenue recognized on an accelerated basis in Q4 2025
Investor presentation filed as an exhibit to the 8-K; the figure is approximate in the filing.
Read the filing on sec.gov, read 2026-10-22
Getty Images also licenses the use of its visual assets and associated metadata to customers in connection with the development of artificial intelligence and machine learning tools.
Read the filing on sec.gov, read 2026-10-22
In August 2025, we launched an AI Training program which enables U.S. subscribers who opted in to monetize their GoPro cloud-based content for AI model training.
Read the filing on sec.gov, read 2026-10-22
Q2 2026 subscription and service revenue included $2 million generated from GoPro’s AI content licensing program.
$2 million, Q2 2026 revenue from the AI content licensing program: Q2 2026 subscription and service revenue included $2 million generated from GoPro’s AI content licensing program.
Read the filing on sec.gov, read 2026-10-22
We also license content for AI which includes training large language models (LLM) at a fixed transaction price and the revenue is recognized at a point in time.
$49.1 million, AI license revenue, fiscal year ended April 30, 2026 (includes $19.4 million from content licensed from other publishers): AI license revenue was $49.1 million in the year ended April 30, 2026 compared to $40 million in the prior year.
Read the filing on sec.gov, read 2026-10-22
As discussed at the Annual Shareholders’ Meeting in May, we've already introduced elements of this approach, including essential data licensing to LLM providers and premium connector access for paying subscribers.
For use in AI products; the filing does not say the content is used for training.
Read the filing on sec.gov, read 2026-10-22
The Dow Jones segment continues to capitalize on a variety of digital distribution platforms, technologies and business models for these products, including licensing its content for distribution on third-party platforms, which is referred to as off-platform distribution, and for artificial intelligence (“AI”) applications.
For use in AI products; the filing does not say the content is used for training.
Read the filing on sec.gov, read 2026-10-22
Content licensing royalties are earned from our relationship with Apple News+ as well as other content use and distribution relationships, including utilization in large-language models and other artificial intelligence (“AI”) related activities.
For use in AI products; the filing does not say the content is used for training.
Read the filing on sec.gov, read 2026-10-22
The licensing of data for machine learning and AI training purposes is a novel business model without an established track record which makes it difficult to evaluate our future prospects and the risks and challenges we may encounter in seeking to execute on this opportunity.
$203 million, aggregate contract value of data licensing arrangements entered into in January 2024 (terms of two to three years): In January 2024, we entered into certain data licensing arrangements with an aggregate contract value of $203.0 million and terms ranging from two to three years.
Read the filing on sec.gov, read 2026-10-22
Our business model includes collecting, processing, and licensing large-scale video and LiDAR datasets to third parties for model training.
Read the filing on sec.gov, read 2026-10-22
We have seen increased demand for access to our metadata for machine learning and generative artificial intelligence model training.
The filing reports Data, Distribution, and Services revenue of $203,264 thousand for 2025, a line that also includes Giphy and Studios, so it is not AI data revenue alone.
Read the filing on sec.gov, read 2026-10-22
Our Q2-25 results benefited from growth in overall customers and, in particular, big-tech customers who licensed our Digital Data for developing their Generative AI applications and LLMs.
Read the filing on sec.gov, read 2026-10-22
(ii) we will license Pathos a comprehensive de-identified multi-modal dataset for the sole purpose of assisting in the development and training of the Foundation Model under the MSA, (iii) Pathos will pay us data license fees of $200 million over a three-year period
$200 million, data license fees from Pathos over three years, including a $50 million upfront payment: Pathos will pay us data license fees of $200 million over a three-year period
Read the filing on sec.gov, read 2026-10-22
On an organic basis revenues increased 5%, primarily due to higher generative AI related transactional content licensing revenue in the Agency business, as well as a contractual price increase from our news agreement with the Data & Analytics business of LSEG.
Read the filing on sec.gov, read 2026-10-22
A Yelp AI API license allows partners to query and display Yelp content and associated context within a conversational consumer-facing product.
For use in AI products; the filing does not say the content is used for training.
Read the filing on sec.gov, read 2026-10-22
DataSeeds offers ethically sourced and fully rights-cleared multimodal - image, video, and audio - datasets that companies use to build and train their AI models.
Read the filing on sec.gov, read 2026-10-22
For companies that aren't public, the buyer database lists what AI buyers publish they pay owners for company records and code.
disclosures.json · disclosures.csv. Licensed under CC BY 4.0: reuse it, commercially too, if you credit Dataconto with a link to this page. The quotes stay the words of the companies that filed them.
Suggested citation:
Dataconto (2026). Public companies disclosing AI data licensing in SEC filings. Version 2026-10-22. https://dataconto.com/open-data/sec-filings