Prynoid

Most apps say they respect your privacy. Few get tested on it.

@Prynoid checks what products actually do vs what they promise. AI, apps, browsers, SDKs, consent flows. Expert-reviewed evidence.

Building something? Get it tested before someone else does: prynoid.com

On 10 Nov, the EU's privacy board published Italy's €1.7m fine on Wind Tre: after a fake "tech support" call, attackers pulled 365,048 customers' records [2].

They just counted the customerId up by one, about 2 million requests, and the API let them [1].

Check: as a test user, change the ID in one request to the next number. You should get an error, not someone else's record [3].

We opened an email app's settings and found "Consent to data sharing" for AI already switched on. That means your data and the content of your emails go to the app's AI provider unless you find this switch.

Consent you never gave isn't consent. Turning data sharing on by default is a privacy violation.

Check: open your email app's settings, find any AI or data-sharing option, and switch it off.

This week, a GitHub user showed Gemini sometimes pulling whole emails and Drive files into casual chats on personal accounts, without being asked to search them [1].

Google's own help page confirms human reviewers see a subset of that connected data [2].

Check: gemini.google.com > Customize > Connectors > Google Workspace > More options > Disconnect [3].

On 10 Nov, Mozilla published fingerprint data from nearly 25 million Firefox users: a commercial tracker could single out 75% of them without any cookie.

Most fixes did little alone; combined, hiding details like taskbar size and graphics quirks cut that to 46% [1].

Check: in Firefox Settings > Privacy & Security, set Enhanced Tracking Protection to Strict [2].