In the mid 90s similar ideas about paperless offices made rounds in news and corporate circles. More recently, Musk attempted to built a full-autonomous factory for Model 3, but it did not pan out due to many edge use cases that needed human intervention.
Full autonomy, is far harder than we think. We will get there, like we got to 7 nines availability in cloud. However it will take lot of leaps and tumbles on our way there. 2030 timeline seems too optimistic
I hear what you are saying but many companies go to great lengths to mention how they are or will use ai on their quarterly conference calls. Sounds like adoption to me..
Yes, this is correct. What companies discuss on the quarterly earnings call about AI adoption is very different from what people mean when they speak of enterprise AI adoption. "Enterprise AI adoption" usually assumes pervasive adoption of AI throughout an enterprise, not isolated examples of, say, marketing teams using Claude to segment customers or IT teams using ChatGPT to refactor legacy code. While those are useful applications of AI, they are isolated and not systemic.
I hate bullet oints, but this is one of the more effective uses I have seen.
I am retired from teaching, but I would have stolen some of this for my Org Theory courses. Very good explanation of organization reaction to potential large changes in any part of there system.
In the mid 90s similar ideas about paperless offices made rounds in news and corporate circles. More recently, Musk attempted to built a full-autonomous factory for Model 3, but it did not pan out due to many edge use cases that needed human intervention.
Full autonomy, is far harder than we think. We will get there, like we got to 7 nines availability in cloud. However it will take lot of leaps and tumbles on our way there. 2030 timeline seems too optimistic
Yes, I agree that 2030 seems too optimistic for a fully autonomous firm, especially one of any real size or complexity.
I hear what you are saying but many companies go to great lengths to mention how they are or will use ai on their quarterly conference calls. Sounds like adoption to me..
It’s not, though. In practice, it’s limited use of things like Copilot and Claude in specific use cases.
Yes, this is correct. What companies discuss on the quarterly earnings call about AI adoption is very different from what people mean when they speak of enterprise AI adoption. "Enterprise AI adoption" usually assumes pervasive adoption of AI throughout an enterprise, not isolated examples of, say, marketing teams using Claude to segment customers or IT teams using ChatGPT to refactor legacy code. While those are useful applications of AI, they are isolated and not systemic.
Excellent summary. Have we reached the "peak of inflated expectations?"
Minor addition to IV.1.b: Enterprise buyers mostly talk to their peers. It's part of what defines a market segment (self-referencing).
I hate bullet oints, but this is one of the more effective uses I have seen.
I am retired from teaching, but I would have stolen some of this for my Org Theory courses. Very good explanation of organization reaction to potential large changes in any part of there system.