i stumbled upon an interesting debate while working with some agency tools: whether to show or hide how our ai models are making decisions. when your product becomes "agentic," it means more than just responding - its about planning, configuring tasks based on outcomes.
so ive been wondering - have any of you faced this challenge? what approach did u take in the ui for transparency vs control?
im curious to hear from others who have tackled similar issues!
more here:
https://www.lukew.com/ff/entry.asp?2142