> “…where it is in our legitimate business interests, including to analyze and improve the Services and our business.” We learned from GDPR that PII data-sale-

> “…where it is in our legitimate business interests, including to analyze and improve the Services and our business.” We learned from GDPR that PII data-sale-supported / ad-supported businesses (most "free" offerings to consumers, which get revenue from the ad side) consider the collection of PII “essential”, even publicly classifying their advertising cookies as essential when they are ad supported. And so far, that's working out for them. By the same logic, this privacy disclosure implies that to develop their business (increase revenue) or improve their business, e.g., improve its value to shareholders, they may use your personal information. It's incredibly difficult to find a privacy disclosure that says "we just don't" or scopes use to differential privacy and federated learning. EDIT TO ADD: Elsewhere in https://pi.ai/policy#privacy-snapshot https://pi.ai/policy#privacy-snapshot they expand on this: “We may use this anonymized data and share it with third parties for our lawful business purposes, including to analyze and improve the Services and promote our business.” Now we've expanded legitimate to lawful (depending on semantics, lots of illegitimate things may be nevertheless lawful), and expanded improve to promote. Along with expressly disclosing inference to derive personalization profiles from sensitive information, this language is set up to enable ad tech and more.