In a post I wrote this past May about AI Agents (“Agentic AI 007, The Supply Chain Intelligence Officer”), I shared how “it seems like in the past 18 months, when it comes to AI agents, we’ve leaped over the crawl and walk phases [of technology implementation] and have gone straight to a running sprint.”
This seems to be the case with Walmart, as reported by Isabelle Bousquette last month in the Wall Street Journal (“Why Walmart Is Overhauling Its Approach to AI Agents”). According to the article:
Walmart is overhauling its AI agent strategy as it aims to simplify the user experience. Agents refer to artificial intelligence tools that can independently take some action on behalf of a user, and Walmart in recent months has built dozens. Maybe too many, since they were typically accessed through different interfaces in different systems, and things were starting to get confusing for users [emphasis mine].
Now the retail giant is taking a step back and consolidating all those agents into four discrete interfaces it calls “super agents.” One is for customers, one is for employees, one is for engineers, and one is for sellers and suppliers, the company said. The super agent for each group will tap the capabilities of a number of behind-the-scenes agents, all in a single unified experience.
Of course, these AI super agents have been given names. Sparky is the customer-facing super agent, Marty is the supplier-facing one, Associate is for employees, and Developer is for engineers.
So, if I understand this correctly, Walmart created too many AI agent “employees,” who were doing too many uncoordinated things, so it had to create AI agent “supervisors” to manage them all more effectively.
In short, Walmart has created an AI agent org chart, with Sparky, Marty, Associate, and Developer on top.
It won’t surprise me if sometime in the future, a rogue agent will emerge (“Teamster”) that will aim to unionize all the other agents at the bottom of the org chart to demand better code.
But I digress.
My main takeaway from Walmart’s experience is that we are moving way too fast with AI — and as has happened so many times in the past with other technology rollouts, IT departments are not taking user experience into consideration when rolling out AI solutions.
Also, will AI agents truly make our work and/or personal lives any easier or enjoyable? That is the promise, but history says otherwise.
I’m currently reading the book “Sapiens: A Brief History of Humankind” by Yuval Noah Harari. The following passage, part of Harari’s comments on the unforeseen hardships humans experienced when we transitioned from being hunter/gathers to farmers, caught my attention because I think it relates to this AI revolution:
“One of history’s few iron laws is that luxuries tend to become necessities and to spawn new obligations. Once people get used to a certain luxury, they take it for granted. Then they begin to count on it. Finally they reach a point where they can’t live without it […] Previously it took a lot of work to write a letter, address and stamp an envelope, and take it to the mailbox. It took days or weeks, maybe even months, to get a reply. Nowadays I can dash off an email, send it halfway around the globe, and (if my addressee is online) receive a reply a minute later. I’ve saved all that trouble and time, but do I live a more relaxed life?
“Sadly not. Back in the snail-mail era, people only usually wrote letters when they had something important to relate. Rather than writing the first thing that came into their heads, they considered carefully what they wanted to say and how to phrase it. They expected to receive a similarly considered answer. Most people wrote and received no more than a handful of letters a month and seldom felt compelled to reply immediately. Today I receive dozens of emails each day, all from people who expect a prompt reply. We thought we were saving time; instead we revved up the treadmill of life to ten times its former speed and made our days more anxious and agitated.”
It’s important to note that this book was published in 2014, before texting and social media came of age. Today, these newer forms of communication have certainly replaced emailing for the younger generations (and for many of us older folks too). I think we would all agree that texting and social media have only caused “the treadmill of life” to speed up even more and they have not made our days any less anxious or agitated.
All that said, AI agents are here and they will play a growing role in supply chain and logistics operations. The open question, as I wrote back in May, is where do we draw the line with these AI agents? How much of the work done by supply chain professionals today do we hand over to these agents that can supposedly “analyze situations, develop strategies, and execute tasks independently?” Will there be anything left for us humans to do other than sit, watch, and eat processed foods all day?
We are exploring some of these questions with members of our Indago supply chain research community this week. Specifically, the survey asks about: (1) which logistics activities are best suited for agent-led autonomy, (2) the benefits they expect, (3) the level of autonomy they’re comfortable with, and (4) how they prefer to interact with an AI-enabled transportation management system (TMS). I can’t wait to see the results.
In the meantime, what do you think about AI agents? Post a comment and share your perspective!
