AI Content Disclosure

The deliverables we produce for you are generated by artificial intelligence. The copy, the images, the video, the voiceover, and the written analysis in our audits are machine-generated. Three consequences follow, and you should understand all three before you use anything we deliver: purely AI-generated work may not be protected by copyright, AI output can state things that are simply not true, and we do not label our output as AI-generated for you.

This document is part of our Terms of Service.

Essentially every creative deliverable in the platform is machine-generated. The systems we use today:

Providers change as models improve. The current list is maintained in our Privacy Policy, which also describes what each provider receives.

We do not clone voices. Voiceover is synthetic speech from a stock synthetic voice. We do not create a synthetic replica of your voice or of any specific person's voice.

Our Terms assign you everything we hold in your deliverables, and you may use them commercially. But the honest answer to "do I own the copyright?" is: possibly not, and the reason has nothing to do with us.

Under current U.S. Copyright Office guidance, copyright protects works of human authorship. Material produced by a generative AI system in response to a prompt, without sufficient creative human contribution, is not copyrightable, and the Office will refuse to register it. Where a human contributes enough — meaningful selection, arrangement, editing, or modification of the AI output — that human contribution may be protectable, but the underlying machine-generated material still is not. Prompting alone has not been treated as enough.

What that means in practice:

• We cannot warrant that a deliverable is copyrightable, and we do not. If exclusive copyright ownership of a specific asset matters to you — a logo you intend to build a brand on, for example — treat AI output as a starting point and involve a human designer and a lawyer.
• Others may legally be able to use it. If a work is not protected by copyright, you may have no copyright claim against someone who copies it.
• Output is not exclusive to you. Generative models can produce very similar output for other people from similar prompts. Nothing prevents that, and we do not promise uniqueness.
• Trademark is a separate question. Whether a generated name or mark is available and registrable is a trademark analysis we do not perform.
• Provider terms also apply. Each model provider sets its own terms about ownership and permitted use of output. We pass through the rights we have; we cannot grant more.

[REVIEW: confirm the current state of U.S. Copyright Office guidance and case law on AI authorship before publication, and confirm the exact assignment language in the Terms is consistent with "we assign what we hold, and warrant nothing about copyrightability"]

We do not fact-check AI output, and you must review everything before you publish it.

Generative models produce fluent, authoritative-sounding text regardless of whether the underlying claim is true. They invent statistics, misattribute quotes, describe features a business does not have, and state confident conclusions from thin or outdated inputs. This is a known property of the technology, not a malfunction we can engineer away.

This matters most in our audits and reports, which make specific factual assertions about real businesses — including, sometimes, about competitors. Those assertions are machine-generated from publicly available signals such as your website text and third-party business listings. They are not verified by a person before they reach you.

Before you publish, send, or act on anything we deliver, check:

• every factual claim, statistic, price, date, and specification;
• every claim about your services, credentials, licensure, or results;
• every statement about a competitor or another named business;
• names, spellings, contact details, hours, and locations; and
• any claim that would be regulated in your industry — health and treatment outcomes, financial or earnings claims, legal advice, superiority or "best in" claims, testimonials, and before-and-after imagery.

You are the publisher of anything you put out. Advertising law holds the advertiser responsible for the truth of its advertising, and "the AI wrote it" is not a defense.

We want to be precise here, because "human-in-the-loop" is easy to overstate.

Our workflow includes an approval step that holds certain jobs before they run. That step is a spending control, not an editorial review: it exists to stop expensive work from running unintentionally. Where the account holder has the ability to approve, the approval can be given by you rather than by us, and a scheduled process clears pending approvals automatically so that queued work is not stranded.

Nothing in that workflow verifies that a deliverable is accurate, lawful, on-brand, or fit to publish. We do not pre-screen generated content, we do not run automated content moderation over it, and we do not fact-check it. Where a managed package includes a person reviewing work before it goes out, that is a service commitment described in your package or statement of work, and it is separate from the automated approval step described here.

[REVIEW: whether marketing copy elsewhere on the site implies editorial human review that does not exist; that mismatch is the most likely source of a misrepresentation claim]

We do not currently label, watermark, or attach provenance metadata to the deliverables we produce. Files we hand you do not carry an AI-generated marker, a visible watermark, or C2PA content credentials.

That means the responsibility for disclosing AI generation, where disclosure is required, is yours:

• Several social and video platforms require creators to flag synthetic or digitally altered content, and some apply their own labels automatically. If we publish or schedule content to a platform on your behalf, applying that platform's AI-content disclosure setting is your responsibility, and you should confirm it on the post.
• Some jurisdictions require disclosure of AI-generated content in specific contexts, including political advertising and synthetic depictions of real people. Those rules are evolving quickly.
• Endorsements, testimonials, and reviews must reflect real experiences of real people. AI-generated content presented as a genuine customer testimonial is deceptive and is prohibited by our Terms.

[REVIEW: whether any labeling, watermarking, or C2PA provenance is legally required for this business today given where clients publish and which jurisdictions they operate in; and whether the platform should implement it regardless as a risk-reduction measure]

You can upload photographs, and our tools can modify them. Because those uploads frequently include real people — staff, clients, patients, and results imagery — two rules apply without exception:

• You must have consent for every identifiable person in anything you upload, covering both the use you intend and the fact that the image will be processed by third-party AI systems. Health-related and before-and-after imagery generally requires specific written authorization.
• No synthetic depictions of real people without consent. Using our tools to generate or alter imagery or audio that depicts an identifiable person doing or saying something they did not do or say is prohibited by our Terms, and may violate right of publicity, deepfake, and consumer protection laws.

• that a deliverable is accurate, current, or complete;
• that a deliverable is copyrightable, original, or unique to you;
• that a deliverable is free of third-party rights, or cleared for a particular use;
• that a deliverable complies with the advertising policy of any platform, or with the advertising rules of your industry; or
• that re-running the same request will produce the same result. Generation is non-deterministic.

Our full disclaimers and the limits on our liability are in the Terms of Service. If a deliverable does not match the brief you approved, the Refund Policy describes the remedy.

Anneshy Royals Inc. (dba OptimizePlus)
28 Geary St., Suite 650
San Francisco, CA 94108
legal@optimizeplus.agency [REVIEW: confirm this mailbox is provisioned and monitored before publication]