Update article "Privacy and AI Assistants"

[Dries-B]

Hi!

I've understood that Posit Assistant now replaces Positron Assistant. In that case, I would like to suggest that the Privacy and AI Assistants be updated. Does everything that is written in the article also apply to Posit Assistant?

[juliasilge]

Thanks for this question! Blog posts are reflective of the moment they are written, and we don't have a practice of going back to update past blog posts. I believe the major claims there are all still true, however, and don't need updating:

• We do not store or have access to any of the data that you send to your model provider: https://positron.posit.co/assistant-provider-info.html#support-and-terms-of-service
• If your model provider is Posit AI, then of course, we as Posit do have access to that data, although not via Positron but via the Posit AI model provider.
• We still think allowing LLM coding assistants to run arbitrary Python or R code is reasonably safe, and we provide a lot of configuration that can dial the risk up/down depending on your particular situation: https://assistant.posit.co/docs/features/permissions/
• It's very important that you have a relationship with a model provider that your trust, for example, with a zero data retention policy: https://docs.posit.co/posit-ai/user/faq/#privacy-data-storage

Is there a different question you had on the ideas from that blog post?

[Dries-B]

Thank you; it is clear now!