AI Terms & Transparency

AI Features can include Ask RELAVANZ, Article Engine workflows, AI-assisted analysis, prioritization, explanation, and other features identified in the product. RELAVANZ may also use deterministic rules, saved evidence, or provider data without a model call for some product states.

"Input" can include prompts, questions, objectives, Site evidence, connected-provider observations, article context, instructions, or other Customer Content supplied to an AI Feature. "Output" can include analysis, summaries, recommended priorities, content drafts, Article Packages, or proposed actions.

Inputs and outputs are treated as Customer Content for purposes of the Terms to the extent they are supplied by or generated for the customer, subject to third-party rights and applicable law.

Current production Agent and Article Engine model execution uses OpenAI, with different OpenAI model families available for bounded stages. Separate RELAVANZ audit and SEO/GEO analysis paths use Google Gemini for supported classification, summarization, content, competitor, authority, and live-presence analysis.

Installed provider clients are not automatically active production providers. RELAVANZ does not promise that a particular model name, model family, or provider will remain unchanged. A provider/model change that materially affects data handling, retention, training, safety, autonomy, or output characteristics is a Legal Health review signal.

Plan and entitlement rules can affect how much new work an AI feature performs. For example, Free Ask RELAVANZ can answer from saved Site evidence without initiating fresh external research or the full paid Agent workflow. Paid/full Agent execution can use additional model reasoning, memory, approved tools, and evidence available to the Site scope.

AI output may be incorrect, incomplete, outdated, or non-unique. You are responsible for reviewing output before relying on it, publishing it, or using it to change a Site or external system. RELAVANZ's validation/monitoring features can provide evidence about a technical change, but they do not make every AI statement true.

AI output is not legal, medical, financial, tax, employment, credit, or other licensed professional advice. RELAVANZ does not guarantee search rankings, traffic, revenue, indexing, AI-answer inclusion, backlinks, reviews, Merchant eligibility, or another external outcome.

Some RELAVANZ workflows can prepare or perform actions through connected applications, APIs, repositories, or CMS systems. The source of authority is the customer's connected account, verified resource binding, granted capability, current workflow state, and any approval or confirmation required by the product.

• Agent skill output remains proposal/control-plane input; the AI reasoning layer does not receive blanket direct-mutation authority.
• GitHub source changes are designed around dedicated branch/commit/pull-request workflows, not direct writes to the protected base branch.
• Provider or CMS operations are limited to detected/granted capabilities and can fail closed when a scope, permission, binding, approval, or precondition is missing.
• Provider completion is not the same as RELAVANZ live verification.

You are responsible for granting only the access you intend, reviewing requested scopes and permissions, confirming high-impact actions where RELAVANZ requires confirmation, and ensuring that you are authorized to modify the connected resource.

Unless a feature expressly states otherwise and appropriate legal controls are in place, do not submit passwords, secret credentials, full payment-card data, regulated health information, biometric identifiers, children's information, or other highly sensitive data to AI Features.

You represent that you have the rights, permissions, and lawful basis needed to provide Inputs and to instruct RELAVANZ and its providers to process them for the requested feature.

Subject to applicable law, third-party rights, and the provider terms that apply, RELAVANZ does not claim ownership of AI output merely because RELAVANZ generated it for your Workspace. AI output may be similar or identical to output generated for others, and you are responsible for evaluating third-party rights before publication or commercial use.

RELAVANZ does not currently operate a general-purpose foundation-model training pipeline that trains a RELAVANZ model on customer content. We do not make a blanket representation that every external model provider will never retain or use API data; provider treatment depends on the applicable contract, API terms, and configuration and must be verified before a stronger promise is made.

The Acceptable Use Policy applies to AI Features. In particular, do not use AI Features to violate law or rights, bypass security controls, deceptively impersonate others, remove required AI disclosures, make prohibited consequential decisions, or process restricted data contrary to feature instructions.

RELAVANZ identifies Ask RELAVANZ and other AI-facing product areas as AI-enabled functionality. If a future feature creates a separate legal requirement for visible AI interaction disclosure, synthetic-content marking, public-interest content disclosure, or high-risk decision controls, RELAVANZ will route that feature through specialist legal review before relying on ordinary AI Terms alone.

RELAVANZ may change models, model families, routing, or providers. Material changes to model/provider identity, autonomy, input categories, retention/training treatment, output modality, or connected-system authority are monitored as legally relevant code changes.

Questions about RELAVANZ AI Features may be sent to hello@relavanz.com.