Recent AI breakthroughs such as in Machine Learning (ML) and Large Language Models (LLMs) are transforming the business decision-making landscape. Further, AI has empowered robots to make autonomous decisions and execute complex tasks. Thus, AI is becoming an excellent alternative and an upgrade for both human efforts and typical automation systems. So, what is the AI impact on business decisions? And, how can AI optimize our business decision-making processes and drive business performance excellence?

To answer these questions, I’ll look at how AI plays a unique role in supporting decision-making. Without a doubt, this is unlike any other forms of information technology. I’ll also share valuable tips on how businesses should leverage AI to support different forms of decision-making. Lastly, I’ll highlight the competitive advantage and benefits that AI brings to businesses.

“Decision is the spark that ignites action. Until a decision is made nothing happens.”

Wilferd Peterson

AI Impact On Business Decisions: The Machine Learning (ML) Effect.

Undoubtedly, AI is transforming the human decision-making process. Indeed, AI is more and more involved in every step of decision-making and decision support. For instance in the context of AI as well as information technology in general, a decision support process can include the following steps:

AI’s Role in Making Decisions

1. Identify Data Sources. Humans work with AI to determine data sets needed
2. Gather and Collate Data. AI with or without human help, trains on data and organizes it.
3. Analyze Data. Prompted by humans, AI examines data using full range of analytics.
4. Determine Options and Make Recommendations. Based on human guidance, AI provides answers, Insights, and recommendations.
5. Makes Decisions. Humans make decisions, or with human permission, the AI agent makes decisions, and even acts autonomously.

Now, AI, such as Machine Learning (ML) possess unique characteristics within a decision-making process over traditional information technologies. One key difference for Machine Learning is that it undergoes a training phase to prepare for effective decision-making support. Then during the decision-making execution phase, AI continues to assist, or even can act autonomously, during each step of the decision-making process. In the next paragraphs, I’ll describe in detail the role AI plays in each step of the decision-making process. In particular, I’ll focus on machine learning AI’s two phases, Training and Execution, within the five-step decision-making process.

Phase One – Train: Build And Teach The AI Machine Learning (ML) Application.

Here the AI developer configures a set of tools for data collection, synchronization, transformation, and analysis. In many cases for ML applications, software developers will train a neural network per the stakeholders’ requirements. Once the development team trains the AI, they will complete the development of the software application to include creating a user interface. Subsequently, the development team releases the software into production. So in the context of decision support, the AI development team completes the first two steps of the decision support process listed below.

AI Training Phase: Decision Support Steps

• Step 1: Identify Data-Collection Opportunities. In the case of ML AI, the software development team identifies the data set that will be used to train the AI application.
• Step 2: Gather And Collate Data. Again, here the software development team working with AI gathers and collates the data as part of training the AI application.

“An expert is someone who has succeeded in making decisions and judgements simpler through knowing what to pay attention to and what to ignore.”

Edward de Bono

Phase Two – Execute: AI Receives New Input, Processes, And Decisions Are Made.

In the AI execution phase, the AI ML application is now in production. As an example, let’s use ChatGPT as the production AI application. In this case, the ChatGPT application is looking for an input prompt from the user. Once it has the input, it will process the data, and provide a response. From there the human can make a decision on what to do next. However, remember this is a simple example. Indeed, there are increasingly many more examples where AI acts as an agent and is completely autonomous in the decision-making process such as a self-driving car. So in regard to decision-making, this AI execution phase is where AI can either support, augment, or even act autonomously.

AI Execution Phase: Decision Support Steps

• Step 3: AI Analyzes Data And Multiple Decision Alternatives. Here the AI application takes the input from the user’s prompt. Then the AI will evaluate multiple decision alternations and narrow down feasible options based on the criteria provided or inferred in the user prompt.
• Step 4: AI Suggests Options And Provides Recommendations. If programmed or requested in the user prompt, the AI application will come back with one or multiple answers. Or the user could ask the AI to regenerate a response that would more than likely provide other feasible alternatives.
• Step 5: AI / Human Makes Decision To Proceed. In the case of ChatGPT, the user will make the decision on what to do with the ChatGPT output. For more autonomous AI applications such as an autonomous vehicle or an AI agent, the AI would make the decision. Also, if it is autonomous, it would then execute the decision.

“A good decision is based on knowledge and not on numbers.”

Plato

For more on AI’s impact on decision-making, see Comidor’s 5 Applications of Artificial Intelligence in Decision Making. Also, for more on AI decision-making models, see Eric Colson‘s What AI-Driven Decision Making Looks Like (see chart below).

Deciding When To Use AI In Decision-Making. 

When considering the use of AI to enhance business decision-making, it is always important to weigh key factors such as costs, implementation time, and business priorities. In addition to these considerations, here are some general guidelines for determining whether to employ AI and for which purposes. Below, we will look at where there are opportunities for fully automated AI decision-making (autonomous), decision augmentation or decision support.

“Whenever you see a successful business, someone once made a courageous decision.”

Peter F Drucker

1. AI Decision Automation Opportunities. 

Decision support automation is usually used to automate simple business processes or tasks that can be completely autonomous. So the question here is can the AI solution do better than a human or other automation. Also as AI advances so quickly, it is able to take on a full range of tasks from simple, complex, and even actual decision-making in chaotic situations. It is really becomes a risk or risk mitigation implementation decision for businesses to use AI or not in any given situation. 

“Whenever you’re making an important decision, first ask if it gets you closer to your goals or farther away. If the answer is closer, pull the trigger. If it’s farther away, make a different choice. Conscious choice making is a critical step in making your dreams a reality.”

Jillian Michaels

The New AI Agent Business Specialist: You Need to Know Their Skill, Expertise, and Comparative Performance

For example, autonomous cars are powered by AI and the question to implement is a matter of risk. Today, car manufacturers are implementing autonomous features in their cars such as self-parking, but the risk is too great for fully autonomous operation in many cases. On the other hand, there are many autonomous pilot projects as well as on-going enhancements to AI that will eventually mitigate most risks involved with autonomous vehicles. Additionally AI software developers can also create AI agents that will work with AI software like ChatGPT to automate the decision-making process. For more on autonomous AI agents, see my article, The New AI Agent Business Specialist: You Need to Know Their Skill, Expertise, and Comparative Performance.

“A real decision is measured by the fact that you’ve taken a new action. If there’s no action, you haven’t truly decided.”

Anthony Robbins

2. AI Decision Augmentation Opportunities.

 Decision augmentation is when a system recommends a decision, or multiple decision alternatives, to humans. Normally, the AI is using prescriptive or predictive analytics. So with decision augmentation, AI coupled with data analytics offers unlimited opportunities. Here the AI does all the heavy lifting, crunches the data, and comes up with one or more options to a business problem. Then the human can choose the option or develop a new course of action and implement it. 

For several decades, AI expert systems have been used in fields such as medical diagnostics to augment decision-making. As AI advances towards artificial general intelligence (AGI), the range of decisions it can augment expands. For instance, with ChatGPT, users can input a problem statement on any topic and receive instant, relevant suggestions. At the same time, AI, such as a LLM, does have many weaknesses to include hallucinations. So, just as with any emerging technology, businesses must weigh the advantages over the risks. To conclude, AI augmentation offers limitless opportunities for decision-makers to leverage AI. Further, AI is able to augment decision-making regardless of whether the business environment or problem is simple, complex, or chaotic.

“We may think that our decisions are guided purely by logic and rationality, but our emotions always play a role in our good decision making process.”

Salma Stockdale

3. AI Decision Support Opportunities.

With AI decision support, humans make the decision, supported by descriptive, diagnostic predictive, or prescriptive analytics. Here humans use AI as a tool to assist with decision-making. So it is not embedded in the decision process like with decision automation or augmentation. Indeed, AI offers countless opportunities to support business decisions. For example, AI can provide decision support in X-ray diagnostics and it just keeps getting better.

So in the case of decision support, AI is not going to replace humans in decision-making. However, it provides unlimited opportunities to improve it. Humans are still needed in business decision making. For example, this quote from Eric Colson sums it up how critical humans are for business decision-making.

“…There are many business decisions that depend on more than just structured data. Vision statements, company strategies, corporate values, market dynamics all are examples of information that is only available in our minds and transmitted through culture and other forms of non-digital communication. This information is inaccessible to AI and extremely relevant to business decisions.

For example, AI may objectively determine the right inventory levels in order to maximize profits. However, in a competitive environment a company may opt for higher inventory levels in order to provide a better customer experience, even at the expense of profits.”

Eric Colson, What AI-Driven Decision Making Looks Like

AI Impact On Business Decisions: Benefits.

So to recap, below are the major benefits that AI brings to Business Decision-Making.

1. Increases Efficiencies.

As with all automation, AI can save time and money by automating and optimizing routine processes and tasks. With AI, it has the potential to outperform traditional process automation plus take on more complex tasks to include autonomous, robotics tasks.

2. Enables Faster Decision-Making.

AI coupled with a cloud-based, high-performance IT infrastructure, enable business-decisions to happen much quicker in real-time versus traditional batch or ad hoc processing.

3. Avoids Human Error.

AI once trained for the task will avoid mistakes and “human error”.

4. Real-Time Processing / 24-Hours A Day.

AI does not get tired, does not need breaks, and can provide answers in seconds. This is not the case with traditional automation or humans.

5. Continues To Learn And Get Better.

Advanced AI is able to continue to grow its expertise.

6. Getting Easier for Business Users To Utilize.

You do not need deep technical experience with the latest iterations of AI. In fact, for assisting with business decisions it is much better for the business user to interface directly with the AI. 

7. A Great Equalizer For Knowledge-based Work.

As AI such as ChatGPT is easy to use, it can greatly improve the knowledge-based deliverables for less skilled or experienced workers. Another way to put this is that less skilled workers are able to use the power of AI to greatly increase the quality and timeliness of their knowledge-based deliverables. On the other hand, a skilled worker will benefit using Generative AI, but not as much.

“Poor decision making I think, is the number one cause for most of our mistakes. So to make fewer mistakes means to make better decisions, and to make better decisions you must train yourself to think more clearly.”

Rashard Royster

More References.

For more on the benefits of AI in decision-making, see NIBusinessInfo’s Artificial intelligence in business. Also, for more information on the growing technical domain of decision intelligence, see my article, This Is What Decision Intelligence Technology Is And Know What Its Not.

This Is What Decision Intelligence Technology Is And Know What Its Not

This article is the fourth in my 5-part series on AI Impact On Business Decisions, see links below:

• The Human Problem-Solving Process – Part 1
• Process Automation – Part 2
• Data-Driven Decision-Making – Part 3
• AI Impact on Business Decisions – Opportunities (This Article) – Part 4
• AI Impact on Business Decisions – Limitations – Part 5

Lastly, if you are in the supply chain industry and have a need to supercharge your decision-making cycles, please contact me to discuss next steps. I’m Randy McClure, and I’ve spent many years solving data analytics and decision support problems. As a supply chain tech advisor, I’ve implemented hundreds of successful projects across all transportation modes, working with the data of thousands of shippers, carriers, and 3rd party logistics (3PL) providers.  I specialize in launching new analytics-based strategies, proof-of-concepts and operational pilot projects using emerging technologies and methodologies. To reach me, click here to access my contact form or you can find me on LinkedIn.

For more from SC Tech Insights, see the latest articles on AI and Decision Science.