The Key Differences Between Analytical and Generative AI

“Since OpenAI announced ChatGPT in November of 2022, many business executives have focused their attention on generative AI,” wrote Babson College professor Tom Davenport, and technology and business strategist Peter High in “How Gen AI and Analytical AI Differ — and When to Use Each,” an article published in the Harvard Business Review (HBR) issue of December, 2024. “This relatively new technology set off a frenzy around AI and caused companies to pay attention to it for the first time. This is a positive development, since the technology is powerful and important, and enables many new business possibilities.”

“However, many companies have been using AI for years with less visibility,” the authors added. “Those that have recently discovered generative AI are at risk of overlooking an older and better-established form of AI, which we’ll call analytical AI. This form of AI is by no means obsolete and is still an important resource for the great majority of companies.”

I’ve been closely following the evolution of AI since the 1970s, especially the more recent era of data-centric AI systems based on highly sophisticated models trained with large amounts of information and powerful computer technologies. We were wowed when in 1997 Deep Blue won a celebrated chess match against then reigning champion Gary Kasparov, — one of the earliest and most concrete grand challenges of AI. The 2010s saw increasingly powerful deep learning AI systems surpass human levels of performance in a number of tasks like image and speech recognition, skin and breast cancer detection, and winning at championship-level Go

More recently, the impressive ability of large language models (LLMs) and chatbots to understand our prompts and questions and generate cogent, articulate sentences has given us the illusion that we’re dealing with a well educated, intelligent human, rather than with a sophisticated stochastic parrot that’s been trained with huge amounts of human language but has no human-like understanding of the ideas underlying the sentences it’s putting together.

“While a few applications of AI employ both analytical and generative AI, the two AI approaches are largely separate,” noted Davenport and High. “To make decisions about the relative importance and value of generative AI and analytical AI, organizations must first understand the differences between the two technologies, and the different benefits and risks associated with each. They can then make decisions about which to prioritize under what circumstances based on their strategies, business models, risk tolerance, and other situations.”

Wikipedia defines predictive analytics, aka analytical AI or predictive AI, as a set of statistical techniques, — e.g., data mining, business analytics, machine learning, — “that analyze current and historical facts to make predictions about future or otherwise unknown events.” Over the past few decades, increasingly powerful and inexpensive computing technologies, new algorithms and models, and huge amounts of data on almost any subject have led to major advances and lower costs in predictive analytics.

Analytical AI is essentially a prediction technology. Prediction usually means anticipating what will happen in the future. Our everyday lives are full of predictions, from guessing who will win a baseball pennant or a political election to deciding what to wear based on the expected weather. Companies use data-driven predictions in a wide variety of activities, like estimating the credit-worthiness of a mortgage applicant or selecting the best customer segment at which to target a new product. Government agencies use predictions in many ways, including economic performance and intelligence assessments. As the cost or predictions continues to drop, we’ll be able to use predictions in all kinds of new applications.

On the other hand, generative AI is primarily a content generation technology, whose primary purpose is to use deep learning neural network models to generate high-quality original content, — e.g., text, images, video, audio or software code, — in response to a a user’s natural language prompts.

“Generative AI can produce content that is original and often indistinguishable from human-created content,” wrote Davenport and High. “Analytical AI is designed to perform specific prediction tasks efficiently, such as predicting when a machine needs service, predicting the price a customer will pay, or recommending products based on user preferences — all based on predictive statistical models. GenAI can’t do these things because it doesn’t deal with these types of data.”

How Do Analytical and Generative AI Differ?

The article explains the key differences between analytical and generative AI in a number of areas. Let me list some of these differences.

Purpose

Analytical AI systems are based on statistical machine learning that are designed for specific tasks, such as classification, prediction, or decision-making based on structured data.
Generative AI uses deep learning neural network models to generate new content — such as images, text, music, or programming code — that mimic human creation.

Capabilities

Analytical AI systems are designed to perform prediction tasks efficiently based on statistical methods, such as when a machine needs service, or recommending products based on user preferences.
Generative AI can’t make predictions because it doesn’t deal with statistical data; instead, GenAI is designed to produce original content that’s often indistinguishable from human-created content.

Algorithmic Methods

Analytical AI utilizes a variety of machine learning algorithms and neural network architectures tailored to specific tasks. The AI models are typically trained on historical data and then applied to make real-world predictions.
Generative AI generally employs more complex methods such as transformer architectures, attention mechanisms, and generative adversarial networks to achieve the desired result. Models are typically created by vendors and customized by user companies because they are large, require extensive computational resources and vast amounts of data.

Data

Analytical AI is based on structured data — typically rows and columns of numbers. The most common form of analytical AI, supervised learning, requires that the data being used to train the model has a known and labeled outcome.
Generative AI uses text, voice, images, and other relatively unstructured data sequences to predict other valid data sequences.

Returns on Investment (ROI)

Analytical AI generally provides good economic returns. Its predictions help businesses forecast demand, optimize inventory management, and identify market trends that lead to reduced costs, improved resource allocation, better insights into customer preferences, and increased revenues. In addition, analytical AI is widely used in risk management and fraud detection.
Generative AI provides economic returns by reducing the costs of content creation compared to those of human content creation, by generating unique and engaging content tailored to individual preferences that can help attract and retain customers, and thus lead to higher customer engagement and revenue growth. In addition, generative AI tools can assist creative professionals by providing inspiration, generating ideas, automating repetitive tasks, and improving productivity and creativity which ultimately lead to better products and services.

How Can Companies Strike the Right Balance Between Analytical and Generative AI?

“Companies will need to determine how to allocate management attention, investments, and talent to these two different domains of AI. A primary consideration is how familiar the relevant stakeholders are with the two types of AI.” Generative AI is often the door-opener because it gets non-technical executives and professionals excited about AI and offers few barriers to its use; analytical AI requires a more technical orientation so its primarily used by more experienced data scientists, although GenAI interfaces can make it easier for non-technical people to use simple analytical models. Companies with a large amount of structured data, such as financial services, retail, and telecom, are likely to be fairly familiar with analytical AI.

“Ultimately, we feel that many AI use cases will combine the two approaches,” said Davenport and High in conclusion. “Together they can fuel new strategies and business models, create more data-driven cultures, yield higher levels of productivity, and facilitate better decisions. Without an understanding of their differences, however, organizations risk under-utilizing one or both types to transform their businesses.”

Irving Wladawsky-Berger, Research Affiliate at MIT's Sloan School of Management and at Cybersecurity at MIT Sloan (CAMS) and Fellow of the Initiative on the Digital Economy, of MIT Connection Science, and of the Stanford Digital Economy Lab.

Dr. Wladawsky-Berger retired from IBM in May of 2007 after 37 years with the company, where he was responsible for identifying emerging technologies and marketplace developments that are critical to the future of the IT industry. Irving was also responsible for the university relations office and for the IBM Academy of Technology where he served as Chairman of the Board of Governors. Irving led a number of IBM’s company wide initiatives including the Internet, supercomputing and Linux.

Since retiring from IBM, Irving has been an Adviser on Digital Strategy and Innovation at Citigroup, at HBO and at Mastercard. He’s been writing a weekly blog, irvingwb.com, since 2005. From April of 2012 until July 2020, Irving was a guest columnist for the Wall Street Journal’s CIO Journal.

Irving served on and later became co-chair of the President’s Information Technology Advisory Committee from 1997 to 2001, and was a founding member of the Computer Sciences and Telecommunications Board of the National Research Council in 1986. Irving is a former member of University of Chicago Board of Governors for Argonne National Laboratories, the Board of Overseers for Fermilab, and BP's Technology Advisory Council. He is a Fellow of the American Academy of Arts and Sciences. Having been born in Cuba and coming to the US at the age of 15, he was named 2001 Hispanic Engineer of the Year. Irving has an M.S. and Ph. D. in physics from the University of Chicago.

Irving Wladawsky-BergerJune 9, 2025Predictive AI, Analystical AI