May 2024 | Point of View

The other AI:  Business leaders need to spend more time with machine learning than Generative AI 

How this AI powerhouse can transform your business 

The other AI:  Business leaders need to spend more time with machine learning than Generative AI 

Generative AI is everywhere these days. Headlines praise its potential, introducing it as a magic wand. But initial excitement often gives way to a more nuanced understanding. While GenAI is changing the game in enterprise technology, there’s also an often-underrated technology in the marketplace—one that’s been in existence for decades—with a considerable amount of untapped potential: traditional machine learning.

GenAI captures headlines, but machine learning holds the title of unsung hero. By using data-backed algorithms to execute complex tasks, machine learning can ultimately become an extremely powerful catalyst for change within an organization. 

How is machine learning different from GenAI?  

The excitement around GenAI originally had many organizations frantic to find ways to adopt it. However, instead of viewing GenAI as the one-size-fits-all AI solution, data scientists are discovering that for many use cases, GenAI isn’t the best AI solution.

Enter: Machine Learning.

GenAI is typically used for content creation or summarization, such as text and images, article reviews, or chat functionality that leverages a data set—things that humans are, to a degree, skilled in. While GenAI showcases the creative potential of artificial intelligence, it's important to note that these are based on large language models (LLMs), which is trained on unstructured, text-based data. 

By comparison, ML models are trained on structured, numeric attribute datasets and offer versatility beyond narrow use cases to discern patterns and guide future decision-making, bridging the gaps from human weaknesses. ML is a complementary solution to people where GenAI can be seen as a threat for replacement, rooted in numeric values that can show the value of a business or customer, whether a medical condition is or is not present based on patient data. 

Many of the issues our clients are facing are actually not suitable for GenAI, contrary to the popular belief that this is the “silver bullet” for problem solving. Machine learning, on the other hand, provides business leaders with the means to extract valuable insights from their data, automate tasks, or enhance operational processes. As automation evolves today’s workplace, developing cases for GenAI, machine learning and employees to work together will be how businesses maximize their outcomes. 

When is machine learning a better solution than Generative AI?

Successfully embedding new technologies within an organization starts with choosing the right tools for the job. In most cases—especially when dealing with structured data—machine learning's ability to analyze and pull insights proves far more suited to the task.

By moving past the misconception that GenAI is the ultimate AI solution and recognizing that machine learning offers a broader range of problem-solving abilities, companies can unlock the true potential of AI and make data-driven decisions that move them forward.


This lack of understanding is seen at all levels of an organization. Sometimes, even C-suite executives have a different perception of AI solutions compared to the actual end-users. To bridge this gap, we need to focus on building the right solution for the right client.

Consider a client we worked with, who asked us to bring in data scientists to help create a categorical predication using GenAI. While the general population has become accustomed to hearing the buzzword “GenAI” as a catch-all for all artificial intelligence, this is actually a traditional machine learning build. Once we highlighted the differences and different use cases, and the client understood the broader context (and cost-benefit trade off) of machine learning, the client was receptive to a different approach.

 ML vs. GenAI: What's the difference?

Machine Learning  

Model program trained on large amounts of data to identify patterns and inform future decisions. Business leaders can use such models to unearth insights from their data, automate tasks, or improve processes.  

Generative AI  

Typically used to create content, including text and images, and is programmed toward a narrow use case.  

Generative AI demonstrates creative applications of artificial intelligence, but machine learning models can be used more broadly and flexibly to address a range of business needs.  

Where Traditional Machine Learning Shines  

Traditional machine learning remains a powerful tool for many applications, especially when dealing with real-world data and seeking actionable insights. This is particularly true for tasks involving customer records or financial reports, where data is often well-structured and labeled. 

Where Machine Learning Excels:  

Categorical Predictions: When you need to predict specific outcomes, like whether an event will (or will not) happen in three months or what the most likely action a user will take is, traditional machine learning algorithms perform very well. They can effectively learn patterns from past data to make these clear-cut classifications.  

Specific examples include:  

  • Customers who are likely to churn can be targeted for retention  
  • The next best product recommendation for a given customer 
  • Whether a customer is likely to be high value upon acquisition and should have time invested by an account representative  

Trend Decomposition: You use trend decomposition to understand the drivers of change in key metrics and provide valuable insights to the business. These techniques can provide valuable insights to inform decision-making. 

Machine Learning is a More Proven Technology Today

GenAI has an inherent “wow” factor that gives it a leg up over other technologies. But the widespread sense of wonder that ChatGPT and other prominent LLMs provide with their easy accessibility can be easily swept aside when compared to the undeniable financial outcomes of machine learning.  

For example, one of our clients in the insurance industry was looking to build a tool to help with claim risk assessment. Our machine learning-based solution was able to help them determine where to allocate inspectors and resources based on patterns in claims data. By proactively predicting claim risks and/or trends, organizations can save money in the long run through improved efficiency.  

While GenAI can potentially reduce costs through drafting tasks or language understanding, its applications aren't as directly tied to revenue generation. In contrast, because machine learning solutions focus more on data, they can be demonstrably linked to cost savings, making the return on investment easier to quantify. GenAI, on the other hand, often relies on bigger assumptions. Leaning into these statistical predictions, the client was able to level up their efficiency and reduce costs. This speaks to our overall goal of empowering clients with confidence in their AI investments. 


Strong data analytics and machine learning are underutilized across many companies, and proficiency in these areas will add greater value to your company than GenAI alone. There are abundant opportunities to create real value with tried and tested machine learning technologies — in addition to being innovative with generative AI. Establishing a clear vision and better understanding of the differences – and use cases – for your organization will help position you for greater implementation and return on investment over time. With ML being the foundation of so many solutions that are becoming fundamental to doing business, now is the time to learn, test and scale.

Explore our latest perspectives