Artificial intelligence has already changed our busy personal lives—and if you don’t believe that, just ask Alexa or Siri. It is now increasingly influencing how businesses operate. Enterprise spending on AI rose 55% between 2020 and 2021, according to surveys by Appen.
Mastercard and other payment processors use AI to identify potentially fraudulent transactions within milliseconds. Manufacturers like Boeing use it to predict when an aircraft part is likely to fail, allowing engineers to engage in preventative maintenance.
UPS and other transport companies use machine learning to optimize driver routes, improving on-time delivery while reducing fuel consumption. Nearly one-third of B2B enterprises are already using AI to forecast sales revenue, identify potential leads, or help sales reps personalize pitches to customers, according to Gartner research. The list is long.
But there are key differences between consumer AI applications and enterprise AI tools that drive business decisions and outcomes. To extract the business value that AI can deliver, business leaders must understand the distinguishing characteristics of enterprise AI, identify appropriate use cases, and build AI solutions and experiences that can get the job done.
What is enterprise AI?
Consumer-facing AI applications such as Netflix’s or Amazon’s recommendation engines train their machine learning solutions using large data sets, often involving hundreds of millions of users. They rely on complex and often nontransparent algorithms trained once or periodically on large data sets to recommend programs or products.
Enterprise AI, by contrast, often deals with problems that have more limited data sets to train the AI, such as improving the experiences of employees. The issues it solves are more nuanced. They tend to be specific to an industry or even a single company.
The needs of a utilities company’s customer-service chatbot, for example, will likely not overlap much with those of a healthcare organization’s chatbot. Hence, an enterprise AI system must be able to quickly and efficiently learn to understand the nuances of a given application and industry, even with small quantities of data.
These purpose-built, “low-data learning” AI solutions are typically deployed in tactical settings, such as decision support, and must conform to high standards of robustness, interpretability, and reproducibility to build trust among the decision makers they support.
Enterprise AI applications must often meet strict regulatory requirements, especially if they rely on consumer data, influence hiring decisions, or are deployed in sensitive industries such as healthcare and financial services. Increasingly, these AI systems are expected to be transparent and explainable, so that users can understand how predictions are made or decisions reached.
Three faces of enterprise AI
Modern enterprises typically deploy AI in three operational areas:
1. Process automation
In this case, the AI system analyzes historical data to model the characteristics of a routine decision-making process. It then applies the models to automate decisions or enable self-service. Examples include an HR app that delivers automated answers to user questions about benefits or travel policies, or an AI system that automatically categorizes incoming requests or incidents into the appropriate group or assignment queue for further processing by an agent.
2. Assistive AI
Many enterprise AI systems assist employees in their flow of daily work, enabling them to complete simpler, more repetitive tasks faster and with less effort, so they can focus on more challenging tasks that make better use of their creativity and domain expertise.
In data center operations, for example, AI systems analyze huge volumes of alerts and failure messages from logs of servers, networks, services, and applications to identify a few plausible root causes. A data center administrator can then use that information to determine an appropriate remediation.
3. Automation discovery
Some large enterprises are starting to use AI to identify additional use cases for automation. In this case, an algorithm might look at all the historical processes involved in a particular task, analyze the sequence, and identify more efficient ways of working. Or it might look at all customer-support records over three months, find the common elements, determine the root cause of issues, and automatically feed it to an engineering team working on the next product iteration.
Speaking our language
As enterprise AI matures, we’ll see additional AI-powered experiences that build on these three fundamental categories.
As natural language understanding improves, employees and customers will increasingly interact with companies in their language of choice via email, phone, web portal, or chatbot. The machines will understand intent and context, converse with the user and enable them to find information, get help, or accomplish a given task.
AI will also help improve the productivity of developers.
For example, companies will be able to use AI in a way that combines sentiment analysis with process automation and assistive AI to transform customer or employee service management. In both cases the AI will gauge the sensitivity of the issue, create a ticket, resolve non-critical issues via self-service, and automatically route more sensitive issues to the most appropriate person for resolution.
Enterprise AI is still in its infancy. Already, workflows are becoming more intelligent and automated. Enterprises are improving their ability to forecast business outcomes and respond more rapidly to unexpected events. As enterprise AI matures, these benefits will multiply.