Roughly how much time does your org spend integrating your generative AI tools together? Are you happy with that time investment?

Of our clients who are embracing GenAI analytics, we find that there isn't a rip+replace approach, but a bit more of an iterative (nailing a few core use cases, ensuring accuracy, guardrails) as a pilot or POC before moving it to general availability.

So its maybe 5-10% of total analytics team time (since they serve other ongoing requests, have legacy BI to support, etc.) but more than for a few folks working specifically on GenAI use cases it's more than that.

We are in the Microsoft GCC and are waiting for the release of MS CoPilot for Government.  We are planning to integrate that toolset into our current environment and have been going through prepping our environment with the MS engineers so we are closer to ready when those tools are released.

We have had some other vendors that have integrated tools into their products which has required us to be focused on evaluating the potential impact/risk associated with those changes.  

We've gone the route of doubling down into a single ecosystem for now so we can fully benefit from the opportunity cost of delivering value fast and learning fast. I see us building and optimizing our toolkit as maturing step 

Not enought yet for sure

The integration of generative AI tools involves substantial time and effort from a team of engineers and researchers. While I don't have exact figures, this investment is crucial for refining, improving, and ensuring the reliability of these tools, aligning them with ethical guidelines and advancing their capabilities.