Preparing for AI pays off

tennis-player-ready-for-serve

Is your enterprise really ready for artificial intelligence (AI)? It may well be, at least in terms of the willingness of top decision-makers to pursue the many benefits that AI can provide, which include valuable insights, streamlined processes, faster execution, and lower costs.

But eagerness alone doesn’t prepare an organization to fully leverage AI’s considerable capabilities. The stage must be set in a number of ways, writes Forbes contributor Michael Trachtenberg, who offers “five things you should consider before getting involved with AI.”

First, he says, is to make sure your network infrastructure is up to the AI task. “That means adopting the latest operating systems, firmware and latest apps from vendors,” Trachtenberg advises. “If your environment is made up of Windows 7, Server 2003 on physical devices, VB apps and individual IIS servers for individual apps, then maybe you shouldn’t be focusing on AI.”

The second thing to consider is whether your enterprise is able to provide the fuel without which AI would merely be a bunch of cobweb-covered algorithms: lots and lots of data. For AI to work as advertised, it is essential that enterprises have a strategy — and platform — for data collection. As Trachtenberg notes, “The more data you have, the more accurate the predictions will be.”

Just as network infrastructure should be current, so should enterprise applications. Otherwise, AI will struggle to effectively manage and optimize those apps, thus undermining at least two of AI’s benefits: operational efficiency and cost savings.

Trachtenberg’s fourth recommendation is to “map out business and application work and process flows,” because that’s your AI’s performance blueprint or roadmap.

The last thing Trachtenberg suggests is to start training your enterprise’s existing systems to understand and respond to human vocal commands. “Building in normal speech syntax search into systems or associating metadata into existing data stores for contextual search will pay dividends long before introducing AI,” he says.

I would add a sixth and seventh thing that enterprises should do ahead of embarking on an AI initiative: Conduct a staff skills assessment and develop a plan to prepare employees. In the former case, determine what skills will be necessary to support AI relative to the skills on hand, then take steps to fill any gap.

In the latter case, demystify AI by walking employees through the entire process. This means demonstrating what happens when you change an algorithm or a variable affected by an algorithm, or even building models that enable project leaders to break down the process into understandable parts. Help them begin to reimagine their jobs as AI allows them to focus on higher-value activities. Some will welcome this change, while others may feel threatened. It’s best to sort all that out sooner than later.

While AI can benefit any organization, those enterprises that meticulously lay the groundwork for an AI initiative will see the biggest payoffs.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: