Operating Modern Models Safely
Paul McMillan · OpenAI
Recording coming soon
This session was given at [un]prompted.au 2026. The recording is being published with the rest of the program; leave your email and you will hear when it lands.
[ Abstract ]
You must budget tokens for model safety. The major model providers do it, so should you. This talk will discuss a variety of approaches, including their downsides, and give you the tools to incrementally improve your own environment.
Get the next talks first
New recordings as they land, the call for papers the moment it opens, and the next [un]prompted announced here before anywhere else. Low volume. Unsubscribe anytime.