Blue Planet Studio - stock.adobe

How to determine a network's AI readiness

Experts discussed how organizations should be preparing for AI deployments within their networks. Determining AI readiness is leadership's first step toward AI integration.

Not everyone is ready for AI -- but they should be preparing for it.

Experts at the ONUG Spring 2024 conference all agreed AI has dug its roots into the IT industry, and it's here to stay. AI is nearly unavoidable as it's increasingly implemented into various aspects of IT infrastructure, including enterprise networks. Network readiness for AI isn't something organizations can afford to delay any longer.

In one session, Michael Milligan -- staff vice president for IT governance, risk and compliance at FedEx -- discussed how organizations should prepare for AI and steps they can take to prepare for AI integration. Milligan explained that people, processes and technologies should all be prepared for AI, especially when it comes to risk mitigation and network reliability.

Why invest in AI?

As AI continues to integrate into and influence nearly every aspect of technology, enterprises have started to invest in it and its various capabilities.

"The investment in AI is coming like an avalanche," Milligan said. "Every single board member at every company is asking 'What is our plan?' and 'What is our strategy for AI?'"

According to Milligan, the boom in enterprise AI investments is due to three main reasons.

1. Automation productivity

In some ways, automation serves as a precursor to AI. Like AI, network automation's appeal is the increase in productivity and efficiency it brings, eliminating various time-consuming tasks for network teams. Because automation can handle different workloads, organizations can continue to drive productivity, profitability and margins, Milligan said.

2. Efficiency

One of AI's greatest appeals is it can help increase network efficiency via improved automated processes and aid with decision-making. According to Milligan, many organizations have unknowingly used generative AI, as some third-party technologies already have GenAI capabilities.

3. Customer experience

Milligan emphasized customer experience as one of the most important business elements people often lose sight of. AI can be used to enhance the customer experience, including using automation and patterns to send targeted messages to customers. AI helps enterprises understand customer behavior so organizations can retain old customers and attract new ones, increasing productivity.

What leadership should be aware of

Though AI has many different benefits, AI deployment should not be taken lightly. It's unwise to integrate AI into networks without first understanding the larger implications of what it can do or having a plan for what organizations want it to do. Before introducing AI into its network, an organization's leadership must build strategies around that deployment.

AI will not replace your jobs, but people with AI skills will.
Gene SunChief information security officer at FedEx

To mitigate risk around AI, Milligan suggested that organizations "look at AI readiness around the triad of people, processes and technology."

AI readiness begins when organizations ensure their team members understand the risks around AI and receive continued education as both the technology and policies surrounding it evolve. If organizations are proactive in employee training for AI tools, they might be able to assuage fears that AI will replace human jobs.

"AI will not replace your jobs," said Gene Sun, the chief information security officer at FedEx. "But people with AI skills will."

Organizations should also make sure they have the proper processes in place before deploying AI. Milligan recommended having process controls for information security to drive down risk. Further understanding where, how and who uses the data helps create security controls around sensitive information. Doing so reduces the risk of information leakage.

When it comes to AI tools and platforms, Milligan advocated for evaluating options. With the large amount of data and workloads AI processes, it needs tools and technology to support it. Organizations should evaluate the products they plan to use for both security and usability.

"We want the fast adoption of AI," Milligan said. "We also don't want to do that at the risk of jeopardizing our customers' data as well."

Milligan recommended that team leadership should be aware of the following considerations as they decide where and how to implement AI:

  • Keep out bad actors and fraud. To help mitigate the risks of AI and keep out bad actors, organizations should have comprehensive business continuity and disaster recovery in place to address any incidents.
  • Have accuracy controls around data quality and output. AI can process and put out vast amounts of data. But it isn't always correct or reliable and might contain misinformation.
  • From a cybersecurity perspective, fight AI with AI. Organizations can use AI to gain insight into their networks, as well as for anomaly detection and faster vulnerability remediation.
  • Protect data from external models. "Protecting data starts all the way at the top of the food chain in terms of the contracting process," Milligan said. He recommended organizations ask questions regarding data protection, compliance and governance.
  • Use responsible AI. Organizations should practice responsible AI usage. Using AI ethically helps enterprises adhere to compliance standards and reduce AI bias, enabling information to be interpreted accurately.

Answering these questions is critical for information security and networking, and the answers can show teams how the collected information will be used.

Nicole Viera is an assistant site editor for TechTarget's Networking site. She joined TechTarget as an editor and writer in 2024.

Dig Deeper on Network strategy and planning

Unified Communications
Mobile Computing
Data Center