Recently, as I was opening up my email, I was shown an alert on an old message I had sent ten days prior. The alert, which asked me if I wanted to follow-up on the message, was right on the money. I did, in fact, need to remind my coworker to reply. The event left me pondering on how far artificial intelligence (AI) in business has progressed and what lies ahead.
40 years ago, when I was studying under Professor Alice Agogino as an undergraduate student at Berkeley, I became enthralled with this topic. During one class, Professor Agogino created this basic diagram.
In those days, AI systems were known as expert systems. They made it possible for a larger range of people to complete more expert-level tasks. They increased the potential of users and changed the business's performance curve. By broadening the group of people who could complete certain tasks, experts were given the extra bandwidth to focus on projects requiring their unique skillsets.
This intriguing idea — that technology can increase human performance at scale — has continued to resonate with me over the years. It has shaped my interests, and my career path has followed the evolution of AI itself.
However, there were a few roadblocks along the way. Our systems' models tended to lag changes in the business; while they solved yesterday's problem perfectly, they eventually became stale. They required dedicated staff for both use and maintenance.
The Current State of AI
In three aspects, today's AI systems differ from earlier expert systems and traditional commercial solutions:
- We employ them in various ways.
- The focus on intelligence has turned away from automation in favor of augmentation.
- AI technology systems develop from usage and adapt to changing business conditions.
How We Employ Them
Typically, traditional business solutions are fragmented. Each system plays a different role, creating a disjointed path forward. One set of systems informs you about how your business is doing right now (conventional business intelligence). Another assists you in making decisions about how to manage your business. Yet another allows you to keep track of what you did to run your business (enterprise resource planning).
Users must travel a long and winding route, from diagnostic and descriptive analytics to predictive analysis and optimization utilizing digital twin models. Then, once they’ve decided how to proceed, they must go back and enter it in their ERP.
Conversely, when using AI, the process starts with recommendations. Users can then investigate predictive, diagnostic, and descriptive insights as explanations. Even if humans make decisions that differ from the recommendations, AI provides step-by-step instructions for taking action.
In the past, expert systems were unduly focused on automation. AI systems can now help users navigate and orchestrate corporate processes. In my opinion, augmented intelligence is a better term for AI than artificial intelligence. Today, we think of AI as a tool that we can train and adjust to our own needs.
Learning & Adapting From Usage
Traditional business systems are stagnant and prone to going stale. They require enhancements and updates in order to implement any business changes or feedback. AI systems, on the other hand, are dynamic. They are always learning and adapting to changing company requirements. They become smarter and more effective over time as you use them more.
What's in Store
These capabilities, while amazing, do come with some challenges.
To begin, getting the most out of AI necessitates careful change management, both in terms of behavior and attitude toward technology. People will try to sabotage technology's progress if they regard it as a danger or a tool by which their performance targets may be stretched. AI adoption will also fail if it results in a loss of control or personal touch with consumers and suppliers, or obstructs cross-functional thinking.
People will, however, work to make AI a success if they perceive it as a new set of tools that makes their life easier, makes them smarter, collaborates across businesses, and achieves more.
Next, we need to consider the ethics of AI. Machine-learning algorithms pick up on what they've been taught and exposed to. If the data is skewed, the AI will be, as well. Furthermore, no one wants to follow instructions from a black box. AI models must be honest about their suggestions, including what assumptions were made, which trends were detected, and which options were investigated.
Finally, we must consider whether the problems we're attempting to solve using AI are the correct ones to begin with. Rather than managing AI, we should use it to expand the human reach and enhance lives.
AI has undergone a dramatic metamorphosis. Our businesses will improve the more we accept, profit on, and improve this technology.
LevaData's AI-powered Cognitive Supply Management platform offers the resources, insights, and confidence needed to turn sourcing into a competitive advantage. Learn more about our integrated platform to discover how you can harness the power of AI for your enterprise today.
Post by: Adeel Najmi
- The Power of Data AggregationAre you paying more than your competitors or peers for the same component? If so, which negotiation strategies—or alternative sourcing […]
- The Role of Data in Sourcing SuccessIt's difficult to source and buy direct materials at scale. While simple transacting may seem easy enough to understand, the […]
- Protecting BOM Health & NPI With Portfolio MonitoringOrganizations are experiencing an increase in unplanned product redesigns, leading to missed product launches and missed shipments of sustaining products. […]
- Lessons for Navigating a Post-Pandemic Supply ChainTwo years ago, I wrote about what I had learned from supply chain executives trying to manage business continuity in the early […]