Hi there,
Consider the use of AI “Guardrails” an important aspect for increased “Rovo” / A.I. adoption in regulated geographies and sectors.
Is the creation of greater Rovo controls being looked at, for example:
- Proxy the underlying LLM provider, to enable (customer designed) guardrails to be inserted (pro-active vs. re-active monitoring / intervention)?
- No doubt Atlassian does some of this for Rovo already, but customers may want their own guardrails applied.
- Provide an administration method for hooking up such a proxy.
Many thanks,
Justin