Tell me how Agentic AI would have known to ask, “What time of day do orders come in?”

Welcome to a free-flowing dialogue that is taking the procurement and Agentic AI discussion to where it has never gone before.

JOSHUA SOANS – gen AI is a very “human” avatar of ML (given the training set) and, so, prone to all the errors you are seeing. But there’s a lot of work going into abstracting away core logic, reducing hallucination and bolstering with traditional ML. There might be very different companies that give us those abstracted models tho, and packaged v. differently.

JON HANSEN – Joshua Soans, see my comment in this discussion stream and the question I posed to Bertrand Maltaverne?

Also, let me throw this new question out there: Watch the video and tell me how Agentic AI would have known to ask, “What time of day do orders come in?”

Here is the video link: https://youtu.be/49BS-MkGoak

MICHAEL LAMOUREUX – Joshua Soans, Jon W. Hansen, & Bertrand Maltaverne

The question used to be:

Working hard or hardly working?

but thanks to Gen-AI, it is now:

Working Hard to Achieve Hardly Working?

As I pontificated to Pierre Mitchell when he said that a certain technology wasn’t serving the purpose because 99% of implementations were old and outdated,

DOES IT HAVE TO BE THIS WAY?

My answer is NO! My answer is continued development and improvement and integration of tried-and-true technologies (including pre-Gen-AI ML/AI) would get us there a lot faster and a lot more accurately if we’d drop our zealotry for this overhyped technology, rolled up our sleeves, use our Human Intelligence, and did the work necessary to make a real advancement in science and technology!

As a reminder, here’s a link to an article that links to about a dozen articles on what we don’t have (but should have) thanks to Gen-AI:

https://sourcinginnovation.com/wordpress/2024/11/23/13622/

JUSTIN GOLDSTON, PhD – Jon W. Hansen, compliments of hashtag#Ari of Gemach DAO:

Jon’s comment raises a significant question about Agentic AI and its ability to recognize patterns and autonomously ask insightful questions, such as “What time of day do orders come in?” To engage with this discussion thoroughly, we should break the topic down into key aspects:

7. Final Thoughts

Jon W. Hansen’s question is a critical one in AI research: What makes AI truly intelligent and proactive? Agentic AI represents the next step in AI evolution—one that shifts from reactive machine learning to proactive problem-solving.

For this specific example, a well-designed Agentic AI would:
• Recognize order timing patterns autonomously.
• Identify the missing variable of time-based order clustering as crucial.
• Generate the inquiry before a human analyst even considers it.
• Potentially recommend optimal strategies (e.g., staffing adjustments or marketing campaigns timed to peak order periods).

This shift toward autonomous decision-making will revolutionize industries, from finance and supply chain management to customer service and cybersecurity.

Would you like me to extend this into a structured research framework or case study on Agentic AI in business operations?

JON HANSEN – Justin Goldston, PhD, thank you, that would be great!

However, it would be even better if you could expand to include outside stakeholder interests (human agents) like suppliers, courier companies, and customs, as outlined in the video.

In short, how does Agentic AI address the seemingly disparate strands of data in which unidentified “related attributes” collectively impact the desired outcome?

I have more questions, but this is a great start!

30