Amid a Pandemic, Where are IT Automation and AI Headed in 2021?

As this unusual year draws to a close, experts discuss how the role of IT professionals continues to change due to AI and automation – and explain what's on the horizon for 2021.

Terri Coles, Contributor

October 16, 2020

6 Min Read
IT automation ai
Getty Images

IT professionals have seen their jobs shift significantly in recent years due to a variety of factors, including an increasing amount of IT automation and AI, but perhaps none arrived as suddenly as the shift to remote working across industries last March thanks to the global COVID-19 pandemic.

As the pandemic continues into 2021 and companies continue to delay re-entry to the office, roll out hybrid workplace strategies or encourage permanent telecommuting, it is hard to predict what the next year will look like in the industry. However, professionals speaking at Interop Digital 2020 did their best to offer guidance for the coming months.

One clear theme emerged: IT automation and AI – including machine learning – will continue to be significant factors for IT professionals across sectors. Here are some ways that will play out in 2021.

Continuing Skilled Worker Shortage

Part of making the right moves in automation and data is putting together the right team of IT professionals. However, the ongoing shortage of IT talent can make that difficult. IT jobs are projected to grow by 12% over the next decade. While computer science graduation rates have grown by 10% over the last decade, that’s still well behind several other technical industries. 

The gap between industry demand and available workers is exacerbated by factors like bachelor’s degree hiring minimums, experience requirements, H-1B visa uncertainty and a retiring workforce, said Damien Howard, executive vice president of social ventures for the technology training nonprofit organization Per Scholas

Computer science unemployment has consistently been about half the overall U.S. rate over the past decade, Howard said, and has stayed relatively low even during the pandemic. Yet only 59% of computer science grads work in a job closely related to their degree, according to data from the Bureau of Labor Statistics cited by Leslie Deutsch, director of learning services at Teksystems Global Services.

“Since 2010, the IT talent market has continued to tighten, meaning that talent is more and more difficult to find,” Deutsch said. 

The end result? An industry with pent-up demand at a time of significant shifts in operations, both short-term (the pandemic) and long-term (the move toward IT automation and AI).

“The demands on organizations to keep pace with technological advancements are higher than ever,” Howard said. COVID-19 has only increased those demands, he said — many digital transformation projects in pipelines became even more urgent with the switch to remote work.

“The demands on IT professions will only continue to grow as companies adjust,” Howard said. That means skilled IT talent will be more important than ever. 

Putting Machine Data to Use

There’s potential in the enterprise for data and machine learning – so long as it’s the right kind of data. There are three types of data used in machine learning, said Andi Mann, chief technology advocate at Splunk. Relational data is “rows and columns,” Mann said — spreadsheets and databases. Reference data involves semantic data like emails and documents. And the third type, machine data, is generated by systems including servers, switches and networks.

This third type of data is where great potential lies for the enterprise space. “If you can read machine data, you can read the whole story,” Mann said — that’s how you follow a customer or client interaction from start to finish, or see the full picture of employee outputs.

In his Interop Digital session, Mann outlined some of the insights that enterprises can reveal via machine data. This data will be increasingly important in 2021 automation and AI efforts as more is done digitally. That includes, for example, employees doing work remotely that might previously be done in person or clients using websites and apps for purchases and other interactions.

For example, using this data, machine learning can be used to highlight clustered events and make predictions, or to look at current events like purchasing patterns. 

“Once you start to understand your patterns, you can start to do forecasts,” Mann said. He pointed to use cases including the analysis of student data about courses, assignment submissions and grades at Clemson University to reveal patterns that can be used to improve student performance.

Sometimes the value of machine data comes not in the patterns but the anomalies or outliers, Mann said. For example, a digital interaction that didn’t work can point to a customer with an unusual but negative experience.

“This is where all data matters,” Mann said. “One data point can show you an unhappy customer who can then go onto Twitter and get everyone to retweet how much they hate you.”

AI: Hype Or Hope?

Romi Mahajan, CMRO at Quantarium, began his moderation of an Interop session with a straightforward question: Does the constant media and industry invocation of artificial intelligence represent something rhetorical or real?

“Every media outlet is focusing on AI,” said Cal Escue, director at Akvelon. There are marginal positives for some enterprises choosing to pursue IT automation and AI – and real ones for others, but it's difficult to figure out where any one organization or use case falls.

“There is a lot of noise, but I think you can kind of sift through it and just really see who's making the changes,” Escue said.

A lot of the messaging around AI comes down to external pressure for companies to have a “story” around it, said Samir Saluja, CEO of Deriveone. However, that external pressure doesn’t cancel out the very real potential. 

“When I think about AI, basically you're trying to optimize human processes,” Saluja said. That can take two forms: trying to replicate things humans do – such as natural language processing – or trying to do things humans can’t do, like large-scale data analysis. 

Those uses of AI are important and will continue and expand in the coming year. But both Saluja and Escue said it’s important to remember that IT professionals are still necessary. “Supervised AI is exactly that,” Escue said. “It's supervised. You need to train a model specifically.” He gave the example of Amazon’s AI for resume search excluding resumes with “women” in them — humans had to override what the system had learned to do.

“The smart systems are smart in what we train them to do — but only in what we train them to do,” Escue said.

Ultimately, the amount of processes to get to full automation is endless, and all those processes need to work together and with those that are not automated or “smart,” Saluja said. Moving toward that IT automation will continue and will free up IT professionals to do more creative work or learn new skills — but it can’t replace those professionals.

“AI is going to be focused on, right now, incrementally optimizing certain human processes” Saluja said.

Moderator Mahajan summed up the ongoing importance of IT professionals, even in the increasingly remote and ever-automated enterprise environment.

“These are not just silicon questions,” he said. “They're carbon questions — questions about people.”

About the Author

Terri Coles

Contributor

Terri Coles is a freelance reporter based in St. John's, Newfoundland. She has worked for more than 15 years in digital media and communications, with experience in writing, editing, reporting, interviewing, content writing, copywriting, media relations, and social media. In addition to covering artificial intelligence, machine learning, big data, and other topics for IT Pro Today, she writes about health, politics, policy, and trends for several different publications.

Subscribe to the Data Center Knowledge Newsletter
Get analysis and expert insight on the latest in data center business and technology delivered to your inbox daily.

You May Also Like