Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

AI models self-hosted by PostHog #29241

Open
Twixes opened this issue Feb 26, 2025 · 1 comment
Open

AI models self-hosted by PostHog #29241

Twixes opened this issue Feb 26, 2025 · 1 comment
Labels
enhancement New feature or request feature/max-ai

Comments

@Twixes
Copy link
Member

Twixes commented Feb 26, 2025

Feature request

Is your feature request related to a problem?

Some users are blocked from enabling AI features due to security concerns around OpenAI APIs – e.g. ZEN-24513.

Describe the solution you'd like

Seems like the easiest way to eliminate the concerns would be self-hosting a model for most tasks, which could replace OpenAI. The data would stay within PostHog's control, removing a subprocessor. The experience would still be seamless for users.

Of course, proprietary models tend to be more powerful, and self-hosting is an effort.

Describe alternatives you've considered

A user also proposed the ability to provide a path to a custom model self-hosted by that company. I'm afraid that would be a minefield though, with lots of ways for things to go wrong, and no quality assurances from us.

Debug info

@Twixes Twixes added enhancement New feature or request feature/max-ai labels Feb 26, 2025
@pauldambra
Copy link
Member

would love this for replay
we'll like want to process embeddings on top of sessions and skipping the network call for those would be 💯

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request feature/max-ai
Projects
None yet
Development

No branches or pull requests

2 participants