License image collections for AI
Inventory the images, creation history and permissions before discussing AI licensing. Separate original photographs, licensed stock, illustrations and extracted video frames.
On this page
Describe the collection
An image collection needs more than a file count. Record subjects, creation dates, resolution, formats and whether captions or labels are available. Distinguish original files from resized copies, crops and near-duplicates.
Sieve includes images in its existing-content program. Defined.ai's partner program describes assessing data quality and metadata. Ask which characteristics matter for the proposed use instead of applying a universal price per image.
Keep provenance with each group
Record who created the images, how your organization obtained them and which agreements apply. Separate commissioned work, employee work, licensed stock and third-party submissions. Note whether any collection has already been licensed elsewhere.
List relevant releases, client restrictions and limits on reuse. A file available on a company drive may have been licensed for a narrow purpose. The rights guide helps organize the review before samples leave the organization.
Make an assessment summary
Describe the collection without including sensitive images. Give measured counts where available, explain the counting method and label estimates. Record missing metadata and known duplicate sets so a later assessment does not confuse total files with distinct usable images.
Ask the program for its sample process, permitted sample uses and retention terms. Agree which files can enter the sample and who may view them. Keep the approved selection separate from the rest of the archive.
Compare the proposed use
Ask whether the license covers training, evaluation, redistribution or products derived from the images. Record any exclusions by subject, collection or recipient. If new photography is requested, separate the assignment and its acceptance terms from licensing the existing archive.
Start with Sieve or Defined.ai where their published formats fit, then use the offer worksheet to compare the written scope and payment. Return to the media overview when the collection includes audio, video or other linked files.
Sources and limitations
Partnership Programs | Defined.ai Checked 2026-11-08
Defined.ai · Primary source
quality, volume and organization, and available metadata … no fixed minimum
Assessment depends on the dataset and intended use; the absence of a fixed minimum does not promise acceptance. Publication date not established.
Partnership Programs | Defined.ai Checked 2026-11-08
Defined.ai · Primary source
proper consent, documentation, and the right to commercialize the data for AI training
The program states these requirements; they do not establish permission for a particular dataset. Publication date not established.
Partner with Sieve Checked 2026-11-08
Sieve · Primary source
licensing content you already own, or by recording the skilled work your team already does. … Every delivery gets quality and rights checks.
Existing content and newly recorded work are separate discussions; the page does not establish one set of payment terms for both. Publication date not established.