Debunking 5 artificial intelligence myths

Since the emergence of ChatGPT, artificial intelligence’s (AI) transformative and disruptive potential has come to dominate news headlines and conversations everywhere. We checked in with our resident AI scholar, Ravi Bapna, to discuss misconceptions and facts surrounding this emerging technology and its related use in business analytics. 

New to business analytics? Get a primer on its historical and future evolution as you learn about what business analytics is and why it’s important. 

Here are five common ideas about artificial intelligence. Are they myth or fact-based?

1. Artificial intelligence is ChatGPT.

Myth.

One prevalent misconception is equating artificial intelligence with specific AI models like ChatGPT. While ChatGPT is indeed an impressive large language model developed by OpenAI, it represents just one facet of the broader field of artificial intelligence. AI encompasses a wide range of technologies and applications, including traditional machine learning involving prediction and description, casual analytics, natural language processing, and robotics among others.

(For an article about AI, we decided to have a little fun and use AI to help us answer some questions. So this answer was provided by ChatGPT and fact-checked by human subject matter experts.)

2. AI, machine learning, and deep learning are all the same thing.

Myth.

Another myth is the conflation of artificial intelligence, machine learning, and deep learning. While these terms are related, they refer to distinct concepts within the field of AI.

What is artificial intelligence?

Artificial intelligence is a general term referring to the discipline of creating machines or programs that use data to solve problems or make decisions like humans.

We’ve all been using AI technology for some time now. Here are some you might be familiar with:

• Web search engines like Google Search
• Algorithm-based recommendation systems like Netflix, TikTok, Spotify, and YouTube
• Human speech assistants like Siri and Alexa

What is machine learning?

Machine learning is a way to use AI. It lets a computer learn without providing explicit instructions on how. Machine learning can use data to:

• Explain what happened (descriptive)
• Predict what will happen (predictive)
• Suggest an action (prescriptive)

This method usually requires human monitoring and intervention. For example, humans are needed to manually label datasets used for training, and to write the machine learning algorithms to process the datasets.

Streaming services, like Spotify and Netflix, use machine learning to learn about your preferences and make personalized recommendations. Each time you indicate you like a song or movie, you give the computer more data to make better inferences about what to recommend to you.

What is deep learning (and neural networks)?

Before we get to deep learning, we have to discuss neural networks. Neural networks is a specific category of machine learning algorithms. It’s modeled on how the human brain uses neurons to process rich sensory data (images, video, voice recording, etc.) and come to conclusions. Neural networks are made of node layers: an input layer, one or more hidden layers, and an output layer. Neural networks made of more than three layers of nodes is considered deep learning.

Deep learning lets machine learning algorithms learn and improve on their own. This decreases the need for human monitoring and intervention. For example, humans don’t need to explicitly define input predictors or features of a predictive model, rather the algorithms can work with the raw data. Reducing the amount of human labor needed would make it possible to scale machine learning.

Although we haven’t seen scaled use of the tool yet, you’re probably more familiar with it than you realize. Deep learning models are also the basis for things like Gmail’s Smart Complete and generative AI algorithms like ChatGPT.

3. Artificial intelligence systems are “black boxes” or impossible to understand.

Myth. 

As the Carlson School’s professor Ravi Bapna likes to say, “Everyone is welcome in the House of AI.”

Developed by Bapna and his colleague Anindya Ghose, the House of AI is a framework and visual representation of the different dimensions that make up this technology. It can help us more easily understand what AI is, how it works, and how individuals and societies can use it to solve problems.

On the bottom floor is the foundation of all AI—data engineering. This is where data is cleansed, aggregated, integrated, and transformed for analysis. Next come four pillars of AI that make up how the data can be used:

• Descriptive analysis — answers “What patterns exist?”
• Predictive analysis — answers “What will happen next?”
• Causal — answers “Did x truly cause y?”
• Prescriptive — answers “How should we respond?”

These pillars support more advanced AI concepts like deep learning, reinforcement learning, and generative AI. Finally, over all the technical aspects of AI are practices and values organizations and society must adopt to implement ethical, equitable, explainable, and fair AI.

While AI systems seem complex, they can be understood. Books for a general audience like Professor Ravi Bapna’s forthcoming Thrive: Maximizing Well-Being in an Age of AI, can be great a starting point. If you’re interested in pursuing AI-related careers, in which you’d work closely with these tools, there are many training and degree programs like the Carlson School’s business analytics master’s program.

4. Artificial intelligence will make human labor obsolete.

It depends.

One of the most pervasive myths surrounding artificial intelligence is the fear that it will replace human workers entirely. While AI has the potential to automate certain tasks and processes, it is unlikely to make human labor obsolete. Instead, AI is more accurately viewed as a tool that can augment human capabilities, enhancing productivity and efficiency in various industries. By automating repetitive tasks, AI allows humans to focus on higher-level activities that require creativity, critical thinking, and emotional intelligence.

(The above portion of this answer was written by ChatGPT. We thought it was a decent response but want to further supplement with thoughts from human subject matter experts.)

At the time of this writing, AI is only good at completing one small task at a time. Most of us are working on multiple tasks that also require judgment. While AI tools can eliminate certain tasks and processes, they can’t automate all aspects of our human labor—especially when it comes to innovation.

For example, analytics requires the parsing of data sources and complex datasets to draw insights. AI-powered tools can aid this process, but human analysts are still needed for things like:

• Data quality checks
• Algorithm bias mitigation
• Insight presentations to stakeholders

5. Leveraging AI for business use requires leaders that understand AI and an AI-ready workforce.

Truth!

According to Professor Bapna, implementing AI to create business value requires “clarity around why AI needs to be used, what AI’s use cases are, and how it can augment human capabilities.” Using AI to its full business potential requires leaders who can envision possibilities, managers who can define opportunities for AI-powered solutions, and data professionals who can use the technology.

And these demands are already here. Based on insights gathered from employer surveys and the Carlson School’s business analytics program advisory board and alumni, there’s a growing need for employees who are proficient in training, implementing, and managing AI models. With the global AI market size expected to have an annual growth rate of 40.2 percent from 2021 to 2028, artificial intelligence literacy will likely be a key differentiator for future analytics professionals and leaders.

How the Carlson School’s business analytics program can help you gain AI skills

In the Carlson School’s Master of Science in Business Analytics (MSBA) program, you’ll develop a good understanding of AI technologies and the ability to harness them responsibly and effectively to create business value.

You do not need any AI experience before the program. In one year, we’ll help you build the foundation you need to understand and deploy advanced AI techniques. You’ll also get hands-on experience training AI models with faculty oversight and support.

We offer an Artificial Intelligence in Business track if you’re interested in enhancing your skills and knowledge in this domain. No matter how much you choose to focus on AI when you’re in our MSBA program, you’ll gain valuable analytics skills and business acumen that will enhance your career prospects.

MSBA Program Snapshot

1 year curriculum (September to following August)

STEM-designated degree 

Merit-based scholarships up to $20,000 available