People talk about the "opt-out" button in the settings and assume this will be enough. But even the text on the toggle implies this is not the full story.
ChatGPT: "Improve the model for everyone" (But are they free to use my data internally, i.e. "not for everyone")
Claude: "Help improve our AI models" (When you click Learn More [0], they admit they may use them to train internal security models etc. if their security classifiers they run on every conversation flag us [with what criteria])
On top of that Codex and Claude Code have different security options and basically they can gain access to everything we have. What is an easy way to do a proper setup for consumer subscriptions.
[0] https://privacy.claude.com/en/articles/12109829-how-do-i-change-my-model-improvement-privacy-settings