TERBISBeyond the Prompt
What happens after you write a good prompt. How to actually work with AI in organizations.
Directing AI instead of just prompting it
A new article from MIT Sloan Management Review separates having a conversation with AI from directing it. It turns out to be the next step in exactly the journey we've followed through this series – from words, to maps, to context.
When agents invent their own language – and humans become the eavesdropper who can't understand
AI agents talking to each other sometimes develop their own, more efficient language – one no human can read. It's a real and active area of research. The question is what happens to oversight and governance once communication stops being humanly legible.
The silent exclusion: competence that erodes without anyone closing the door
Part three of 'Beyond the prompt'. First we could no longer read what the agents said to each other. The next step is quieter: we ourselves lose the ability to contribute, not because we're shut out, but because the skill is never practiced.