AI and Robotics: Trends and Opportunities

Artificial intelligence (AI) has been around for decades. But generative AI—the branch that's generating all the buzz right now—is relatively new.

These tools, built on large-language models (LLMs), are trained on the collective knowledge stored on the internet. Because LLMs excel at pattern matching, they can provide answers that seem humanlike. But pattern matching also means they’re only as accurate as the data they collect.

Unless you have a wealth of data available that you can tailor and customize for your needs, these AI tools probably aren’t ready to handle your daily operations. And most companies don’t have the capacity to invest hundreds of millions of dollars to build an LLM that’s specific to their needs. However, several companies offer off-the-shelf generative AI tools, and companies can tie their proprietary data to these existing LLMs.  

Similarly, companies have used robotics in one form or another for decades. The difference now is that AI-powered robots promise to bring transformative capabilities, not just efficiency improvements.

In the near term, the key will be understanding how AI can work for your organization now, and what opportunities you need to prepare for in the next few years.  

How can different industries benefit from AI and robotics? 

Let’s look at two distinct industries: manufacturing and financial services. In the financial industry, a lot of day-to-day processes are based on legacy systems, including paper. Newer AI models can process this unstructured data. That’s particularly important for certain sectors of the financial industry that are dependent on personal relationships, such as real estate or commercial lending. 

Current AI tools can provide new use cases for working with the data gleaned from these relationships. That could include customer service chatbots or recommendation tools for banking relationship managers.

In the financial sector and other industries that depend on human interaction, AI can deliver improved personalization. Whether it’s tailoring a loan for a business owner’s unique needs or crafting personalized emails for better customer retention.

As for manufacturing, companies have used robots since the 1960s. The difference is that robots have typically been programmed to complete specific tasks to make those tasks as efficient as possible. Eventually, robots could be programmed to produce a new type of car, or new type of engine. That would enable an automaker to readapt a manufacturing plant quickly.

In the near term, beyond manufacturing and financial services, AI and robotics will be making inroads into healthcare, with AI systems predicting diseases and robots performing surgeries.  

How can companies position themselves for future AI opportunities? 

According to several reports, we’re already starting to run out of high-quality language data that’s publicly available on the internet, making it a challenge to continually improve the quality of the results AI tools generate. The question is, where do we get the next level of data? 

Companies with strong data-collection capabilities will be in a prime position. Tesla, for example, is collecting data from the sensors and cameras in its cars to generate data sets that can be used for training deep neural networks and improving autonomous-driving capabilities.

Lots of companies generate a wealth of data that could be useful to current generative AI systems, but they're locked inside of these organizations. Given the value of this information, we’re starting to see companies sell their data. Ahead of its recent IPO, Reddit struck a deal to make its content available to train Google’s AI models. Similar deals have been struck by several other companies. But how do you maximize the value of the data your company owns? 

Many of the AI experts we speak with point out that many companies don’t have a good understanding of what they need in terms of how to capture the right data. A large part of that involves having the right talent in place to manage the data in a way that’s useful to an AI model. That's why the successful companies will be the ones that invest in capturing the right data and attracting the right talent to do so.

Surprisingly, the types of companies currently doing this are not always the ones you might expect. You’d probably think high-tech industries like aerospace would be at the forefront, but that’s not necessarily the case. That's partly because the information inside these types of organizations is often trapped in different silos. Industries like retail, on the other hand, are focused on certain objectives, such as improving sales, and that information is more readily available across the organization.

We're in the beginning stages of the AI revolution, which is an exciting time because so many possibilities are ahead of us. The key will be understanding how to position your company to be able to take full advantage of AI’s potential.