Forum Discussion
Does AI work for you or do you work for AI?
Are you a centaur or a reverse centaur?
What does that mean? Here's how Cory Doctorow presents those terms in his book, "The Reverse Centaur's Guide to Life After AI":
"[A] centaur is a person who is assisted by a machine...
A reverse centaur is a human who is conscripted into acting as an assistant to a machine."
As an example, think about Lucy & Ethel working at the chocolate factory. (Search for the video if you don't know what I'm referring to.) They're supposed to wrap each chocolate as it travels on a conveyer belt. But they can't control the speed and can't keep up. The result is hilarious in a situation comedy. It wouldn't be so funny in an actual work situation, though. Product quality would suffer, and so would the workers.
Or consider this note by Ross Stevenson:
- (I recommend going to that LinkedIn post for his full list of takeaways from a conversation about how people learn.)
So the real question is this:
How can L&D pros use the new AI features in Articulate to create, review, and maintain meaningful and accurate "content" that helps improve job performance throughout their company?
Of course, so many other questions have to be answered first. How big is the company? What tasks do people need to do? How often do processes/procedures change? What has to be done to ensure that materials are error free? What are the risks from errors and/or noncompliance? What's the budget? How much input does L&D have in the decision making anyway? And on and on...
In other words, this discussion won't be marked as "solved." It's just here because I think there's a lot worth pondering—and this community is great at that. 😊
To paraphrase Hamlet: "No software is either good or bad, but the use case makes it so."
P.S. For those who might be interested in Doctorow's book, here are more excerpts:
3 Replies
Hey Judy — love this framing. I work on an innovation team as the AI integrator, so I think about this question almost daily.
My answer is centaur, but I'm strict about who's driving. AI does the heavy-lifting prep: first-pass storyboards, placeholder narration so I can feel the timing before paying for real VO, variations of a scene image so I can compare options fast. I stay on: the learning objective, the narrative spine, and every final word and pixel that ships.
The trap I watch for is reverse-centaur creep — when the model's first draft starts anchoring my thinking. If I generate before I've sketched the structure myself, the output quietly narrows what I consider possible. So my rule: I write the design down before I open the model. AI refines my plan; it doesn't write it.
Honestly? AI does the busywork, and it does it fast enough that I can spend the saved hours on the parts that actually make training work — talking to SMEs, watching real learners, and iterating. That's the trade I'm happy with.
It sounds like you've got a good handle on how to use AI—and the awareness to avoid "reverse-centaur creep."
So "Yes!" to making thoughtful decisions about when to use AI. That's a vital part of AI literacy. (Sadly, some workers don't get much choice.)
Fascinated by this take, JudyNollet​. In the community sphere, because our work is so people-centric, there has long been a debate even before the advent of AI about what things should and should not be automated. For example the community industry always argues about the utility of automated "welcome" messages for communities—yes, they scale, but they do so while missing the actual value that getting a warm welcome from a person brings to a community setting.
So I see a lot of the questions you posed here that "must be answered first" as relevant to that ongoing conversation about how we use technology effectively rather than lettting it use us. I appreciate you sharing this (& the Hamlet quote!).
