Cutting Through The Buzz: Four Categories of AI You Need To Know

We hope that after reading this article you can start thinking of these terms less as buzzwords and more as real tools.

When we talk about Artificial Intelligence (AI), we’re actually talking about a host of technologies that are applied in different ways. In each instance, the goal is to gather unstructured data, normalize it, and then use it to produce insights and recommendations. These technologies all fall under the umbrella of AI, and are deployed depending on the needs of the company. For example, some companies will choose to implement a portfolio of modules designed for projects, such as supply chains or project bidding, while other implementations target corporate strategies, such Business Development and Risk. Some companies may target both.

AI was first defined in 1956 at the first-ever summit on the topic at Dartmouth College. A group of 10 scientists convened "to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it." Thus, the broad and now-popular field of AI emerged, the concepts of which span from superhuman general intelligence to email spam filtering. This layout should help to block the noise and narrow the applications of AI to the enterprise world, and how it can help you run a smarter company.

As you read this list, keep in mind there are many overlapping concepts among these tools. Most all of them subscribe to the underlying fundamentals of data science and machine learning, which relies on an algorithm constantly being fed data and taught to detect trends and patterns. Let’s have a look at the tools that comprise “AI”, and how they are being used today.

  • Machine Learning (Deep Learning, Neural Networks)

  • Natural Language Processing

  • Computer vision

  • Robotic Process Automation

Machine Learning

Machine learning is a framework of statistical models that perform actions based on pattern recognition rather than human intervention. These algorithms are manifested in a number of ways, but they all “learn” how to perform some action, whether it is reading documents, automating workflows, or creating predictive analytics on data sets. And all of it is happening without being programmed what to produce.

Natural language processing (NPL)

Whether you’re the most technical in your company or the least, you know that computer code, including Python, Java, or C++, is structured differently than human language. NLP is a layer that gives computers the ability to learn how to interpret and manipulate human language. Siri and Alexa are two of the best NLP examples at the consumer level, but the technology has enormous benefit at the construction enterprise level as well. Imagine typing the request “Find a road project that I have a high chance of winning” into your AI ERP and returning results based on your bid/win ratio and those of your competitors on similar projects.

Computer vision

This is arguably AI’s biggest strength today, and it allows for the easy reading of static PDFs or cut sheets without having to digitize them. Computer vision can recognize letters, shapes, and depth in images, and is particularly useful when an algorithm needs to process hundreds or thousands of documents, such as RFIs, Submittals, Purchase Orders, Invoices, and Punch Lists. It is applied in everything from Apple’s Face ID to Google’s Optical Character Recognition (OCR) technology known as Vision.

Robotic Process Automation (RPA)

This technology targets workflow automation and typically includes back office work that is repetitive. Whereas computer vision can read the hundreds of RFIs, Submittals, and Invoices that are produced in a project, ML-based RPA bots can perform actions using the data extracted from those documents. And as data is read and processed in the RPA process, it is being fed back into the neural network nodes of the AI program to constantly improve and refine its outcomes.

Honorable mentions

While technically not considered artificial intelligence, these tools form the backbone of the data infrastructure needed to render predictions and analysis.

Big Data

With more and more data being create each year, it is humanly impossible to not only read and analyze the data, it is also difficult to make sense of it and make decision based on its conclusions. These AI tools will be the managers of the Big Data Universe.


As more and more data becomes available, it is important to have a storage mechanism that is immutable and secure. Blockchain is a perfect storage mechanism for these complex data sets, and ensures that any instances of project information are properly accounted for.

Briq is working to bring all of these powerful technologies to construction. Sign up for our webinar to learn more!

Ellis Talton