Rezolve Ai Unveils Patent-Backed AI Provenance Platform as Watermarking Becomes Global Infrastructure

• Digital provenance market projected to grow from $4.2 billion in 2026 to $16.9 billion by 2033 as AI marking becomes a legal requirement
  
• End-to-end, model-agnostic stack enables LLM owners and text, voice, image and video generators to mark, sign and verify AI content, with patent-backed image watermarking embedded within the content itself
  

NEW YORK, Aug. 13, 2026 (GLOBE NEWSWIRE) -- Rezolve Ai (NASDAQ: RZLV), a global leader in AI-powered commerce infrastructure, today unveiled Rezolve Provenance, a patent-backed, model-agnostic platform designed to mark, sign and verify AI-generated and AI-modified content across enterprise and commerce workflows. The platform combines signed C2PA Content Credentials, secure provenance records and independent verification with content-embedded watermarking for images, creating a durable trust layer that can operate across different AI models and platforms.

• Complete provider stack: Post-generation marking, signed C2PA Content Credentials, content-embedded image watermarking, secure provenance records, detection and verification APIs, SDKs, private-cloud deployment and TraceWare audit.
• Built for AI content generators: A model-agnostic infrastructure layer for LLM owners and platforms generating text, voice, images and video, integrated after generation without retraining the underlying model.
• Durability beyond metadata: Anthropic says heavy edits can make its text mark undetectable and that file metadata can be stripped. Rezolve's patented image mark is encoded within the visual content, so stripping metadata does not remove it.
  

AI marking becomes a legal requirement
Transparency obligations under Article 50 of the EU AI Act became applicable on August 2, 2026. They require in-scope providers of generative AI systems to mark AI-generated or manipulated audio, image, video and text outputs in a machine-readable format and ensure they are detectable as artificially generated or manipulated, subject to specified exceptions and technical-feasibility considerations. A limited transition to December 2, 2026, applies to certain generative AI systems placed on the market before August 2. Non-compliance can expose companies to fines of up to EUR 15 million or 3 percent of worldwide annual turnover. 
Industry adoption is accelerating alongside regulation. Anthropic announced this week that supported Claude models will embed watermarks in generated text and attach digitally signed C2PA provenance metadata to supported files wherever Claude is offered worldwide. Rezolve Ai believes that commitment by a leading foundation-model company validates both the scale and urgency of the market Rezolve Provenance is designed to address.

“Anthropic's commitment to mark AI-generated content is an inflection point for this market,” said Daniel M. Wagner, Founder, Chairman and CEO of Rezolve Ai. “It confirms that provenance is becoming essential AI infrastructure. The next challenge is durability: Rezolve Provenance adds a model-agnostic second signal embedded within the image itself.”

A $16.9 Billion Digital Provenance Market
Grand View Research estimates that the global digital provenance market will grow from approximately $4.2 billion in 2026 to $16.9 billion by 2033, representing a compound annual growth rate of 21.9 percent¹. Growth is being driven by the rapid expansion of AI-generated content, increasing concern over deepfakes and misinformation, emerging transparency regulation and growing demand for reliable digital-content authentication.

The market encompasses digital watermarking, content authentication, AI-generated content detection, secure provenance records and verification infrastructure. Rezolve Provenance brings these capabilities together within a single end-to-end platform, from marking and signing content at the point of generation to durable binding, independent downstream verification and audit.

Rezolve’s addressable customer base includes foundation-model and LLM owners; specialist text, voice, image and video generators; cloud and enterprise AI providers; device manufacturers; social networks, publishers and advertising platforms; retailers and marketplaces; and verification, compliance and public-sector systems.

An end-to-end provenance stack for AI content generators
Rezolve Provenance gives providers a deployable stack rather than a standalone watermark or metadata label. A single integration can mark and sign content at the point of generation, bind it to a secure provenance record, make it independently detectable and verifiable downstream and audit how AI agents act on the resulting trust signals.

• Mark and sign at creation: Apply machine-readable marking and signed C2PA Content Credentials as output leaves the model or generation service.
• Bind provenance to the content: Encode a patented recoverable signal in image color channels so the asset can reconnect with its provenance when metadata no longer travels with it.
• Register and resolve: Link the mark to secure records covering creator, model or service, licensing status and applicable content or commercial policy.
• Detect and verify anywhere: Use APIs, SDKs and private-cloud deployment to verify content across platforms, marketplaces and enterprise systems.
• Audit downstream use: Use TraceWare to record whether AI agents checked provenance and followed policy before using, distributing or recommending content.
  

The common marking, signing, verification and audit layer is designed for LLM owners and specialist platforms generating text, voice, images and video. Rezolve's patented content-embedded watermark supplies an additional durable binding for images, while the wider stack supports provenance workflows across content formats and can complement provider-native systems.

“LLM owners and content-generation companies do not need another point feature,” Wagner said. “They need the complete control plane: marking, signed provenance, a durable content binding, independent verification and an audit trail. Rezolve brings those layers together in one model-agnostic stack.”

Designed for durability
Anthropic's published limitations identify the durability gap directly: its text mark may not be detectable after content is heavily edited, paraphrased, translated or mixed with other writing, while file metadata can be stripped through conversion, re-saving or screenshots. Rezolve's patented image watermark is embedded in the visual content, not carried only as file metadata. Stripping metadata does not remove the embedded signal.

That embedded signal provides the soft binding recognized by the C2PA architecture: a way to reconnect an asset with its provenance record when metadata no longer travels with it. Rezolve Provenance combines the mark with secure records and independent verification and is designed to maintain recoverability through common cropping, compression, reformatting, screenshots, printing and recapture.

Proven in commerce, now applied to AI provenance
Rezolve's technology is not a laboratory concept. It was originally developed to make images and video instantly interactive, allowing a consumer to move directly from what they saw to a product, service or transaction. An archival demonstration shows the capability operating across printed media, outdoor advertising, television, packaging, merchandise and digital screens [Video demo:
https://rezolve.com/videos/rezolve-ai-unveils-al-provenance-platform-as-watermarking-becomes-global-infrastructure/].

Rezolve is now applying the same core capability to AI provenance: binding content to a verifiable record through a signal embedded within the content itself, rather than relying solely on metadata that can be removed.

The mark can carry or resolve to a creator, generation service, licensing status, content policy, merchant, product or campaign. Applied after generation, it requires no retraining of the underlying model and can complement provider-native systems or operate independently.

U.S. Patent No. 12,412,310, “Encoding Data Matrices into Color Channels of Images Using Neural Networks and Deep Learning,” covers the use of machine learning to encode error-corrected data matrices across the color channels of an image. The patented approach is designed to create an imperceptible, recoverable signal within the visual content itself.

From provenance to an active commerce control
Retailers are already deploying AI-generated lifestyle imagery, personalized advertising, virtual product demonstrations, localized video and synthetic presenters. That content is increasingly becoming an input to the AI agents that search catalogs, compare products and recommend purchases. In that environment, provenance is not simply a label for a consumer to inspect; it can become an active control within the automated commerce workflow.

Rezolve Provenance links an asset to its verifiable source and commercial rules. TraceWare records whether an agent checked those signals and followed policy. Together with brainpowa, they create an integrated commerce trust stack: grounded AI intelligence, verifiable content provenance and auditable agent actions, available through APIs and SDKs.

Commercializing the Opportunity
Rezolve Ai is launching a design-partner program for AI model owners, content-generation platforms, cloud providers and major enterprise and commerce platforms. Commercial models include enterprise licensing, usage-based marking, verification API fees, private-cloud deployment and provider-branded verification services.

Media Contact
Rezolve Ai
Urmee Khan
Global Head of Communications
urmeekhan@rezolve.com
+44 7576 094 040

About Rezolve Ai
Rezolve Ai is a global leader in AI-powered commerce technology. Its Brain Suite platform helps retailers, brands and financial institutions transform how consumers search, engage and transact across digital channels.

Rezolve Ai's proprietary brainpowa™ models are purpose-built for commerce, while TraceWare™ and Auditable AI provide transparency and accountability across agentic AI workflows. Its proprietary distributed database platform provides trusted, real-time data infrastructure for AI agents and enterprise applications. Together, these technologies enable enterprises to deploy AI that can engage customers, understand intent, support transactions and operate safely at scale.

Headquartered in London with operations across North America, Europe and Asia, Rezolve Ai partners with leading global enterprises to power the future of commerce through AI that sells.
Learn more at www.rezolve.com.

Forward-Looking Statements
This press release includes “forward-looking statements” within the meaning of the safe-harbor provisions of the Private Securities Litigation Reform Act of 1995. Words such as “expect,” “estimate,” “project,” “forecast,” “anticipate,” “intend,” “plan,” “may,” “will,” “could,” “should,” “believe,” “predict,” “potential,” “continue” and similar expressions are intended to identify forward-looking statements.
These statements include, without limitation, statements concerning the capabilities, development, availability, durability and performance of Rezolve Provenance, its interoperability with third-party standards and provider-native systems, anticipated customer and partner adoption, the design-partner program, potential commercial models and routes to market, the scope and enforcement of applicable regulatory requirements, the strength and scope of Rezolve Ai's intellectual property, and future growth.
Forward-looking statements involve significant risks and uncertainties. Readers should carefully consider the risks and uncertainties described in the “Risk Factors” section of Rezolve Ai's Annual Report on Form 20-F and its subsequent filings with the U.S. Securities and Exchange Commission.

Rezolve Ai cautions readers not to place undue reliance on forward-looking statements, which speak only as of the date made. Except as required by applicable law, Rezolve Ai undertakes no obligation to update or revise forward-looking statements as a result of new information, future events, changed circumstances or otherwise.
_________________________
¹ Source: Grand View Research, “Digital Provenance Market Size & Share Report, 2026–2033,” June 2026