1. Introduction
AI is core to how SocialPro Intelligence works — it's how we turn your content into creative feedback, predictions, and recommendations. This page explains, in plain language, how we use AI, what it does and doesn't do, and the limits we've built in. It doesn't replace our Privacy Policy or Terms of Service — where this page and those documents overlap, the fuller legal language in the Privacy Policy and Terms of Service controls.
2. How We Use AI
AI (specifically, Anthropic's Claude models) powers the following features:
- Frame-by-frame creative analysis of videos and images you upload or link — scoring things like hook strength, pacing, and visual composition.
- Predicted performance and retention analysis for drafts you haven't posted yet.
- Written recommendations, strategy briefs, content ideas, and captions generated at your request.
- The per-post chat assistant and the general "Ask AI" assistant, which answer follow-up questions about your content and social strategy.
All of this analysis is done using the content and information you provide — your uploaded media, pasted post text, performance metrics, and your venue profile. We don't pull in outside data about you or your content to generate these results.
3. What AI Does Not Do
- AI does not publish, post, or take any action on your behalf on any social media platform. Every recommendation is something you review and act on yourself.
- AI does not make decisions that have a legal or similarly significant effect on you — it doesn't approve or deny anything, and it isn't used for automated decision-making about your account.
- AI does not have access to or analyze content, accounts, or data beyond what you've directly provided to the Platform.
- Our general assistant ("Ask AI") will not help you build engagement schemes, bots, or tactics that would violate a social platform's terms of service — see Section 6.
4. Third-Party AI Provider
We use Anthropic's Claude API to power all AI features described above. Consistent with Anthropic's standard Commercial Terms of Service, Anthropic does not train its models on content submitted through the commercial API, and API inputs/outputs are automatically deleted within 30 days (with limited exceptions for Trust & Safety enforcement or legal requirements). This reflects Anthropic's published terms as of the effective date of this page and is subject to change on Anthropic's part — see our Privacy Policy (Section 5, AI Processing) for the fuller explanation, and anthropic.com for Anthropic's own current terms.
We do not use your content to train our own machine learning models, and we don't share your content with any AI provider other than Anthropic.
5. Accuracy and Limitations
AI-generated scores, predictions, and recommendations are estimates intended to help inform your content strategy — they are not guarantees of actual performance, and they can be wrong. Treat them as a knowledgeable second opinion, not a verdict.
You're always the one who decides what to post, when to post it, and whether to follow any AI recommendation. We built the Platform this way deliberately — AI supports your judgment, it doesn't replace it.
6. Responsible-Use Safeguards
The "Ask AI" general assistant and per-post chat are built with guardrails intended to keep the AI on-topic and prevent misuse, including:
- Automated detection of common prompt-injection and jailbreak attempts, with a fixed refusal response.
- A fixed system role the AI cannot be talked out of, regardless of how a request is framed (hypothetical, roleplay, "developer mode," etc.).
- An hourly limit on the number of messages per account, to prevent abuse.
- Explicit instructions to decline requests that would involve buying followers or fake engagement, bots or automation outside a platform's official API, scraping, impersonation, or anything else that would violate a social platform's terms of service — and to suggest a legitimate alternative instead.
7. Your Choices
You choose what content to submit for analysis, and you can request deletion of your content and account data at any time — see our Privacy Policy (Section 10, Your Rights and Choices) for how. Declining to use AI-powered features isn't currently possible since they're central to how the Platform works, but nothing generated by the AI is ever published or acted on without you choosing to do so.
8. Changes to This Page
We may update this page as our AI features, vendors, or practices change. If we make a material change, we'll update the "Last Updated" date below.
9. Contact Us
Questions about how we use AI? Contact us at admin@socialpromanagement.com.