BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

How To Implement And Scale AI In Your Organization

Following

(This article is part of a series on Artificial Intelligence for Board Members and Senior Executives.)

The jury is in. AI is the key to near-term success in virtually every industry, and top executives know it. 94% of business leaders agree that AI is critical to their organizations’ success over the next five years, and 79% say they have fully deployed three or more AI applications. (This article summarizes the fifth edition of Deloitte’s 'State of AI in the Enterprise’ report.)

But it is one thing to come to a verdict about strategy. It is something else altogether to implement and scale that strategy. Artificial Intelligence brings in new risks that most organizations have never faced before, including concerns about discrimination, unethical use of AI, environmental damage, and the most significant risk of all — permanently falling behind.

Successfully implementing and scaling AI requires cultural change, new job roles and training, and clear commitment from the top. Here's a summary of the four key actions that companies should take to get this right:

1. Invest in leadership and culture

Leaders must find ways to harness optimism for AI within their workforce so that AI becomes a positive part of the organization's culture. When employees feel that AI can help enhance their performance and increase their job satisfaction, they will be more likely to embrace the technology and support the cultural transformation needed for AI's success.

Deloitte also identified agility, willingness to change, and executive leadership committed to change management as critical characteristics to successfully implementing AI. Specifically, it is essential to appoint a leader to help workers collaborate with AI-based solutions.

2. Transform operations for AI integration

Whereas some 82% of respondents reported a positive employee response to AI, things appear to be rocky regarding operations and risk management. Respondents identified AI-related risk management as a top inhibitor of initiating and scaling AI projects. They identified two tactics to help mitigate risk—training developers on AI ethics and training/supporting employees who work with AI.

While those are promising approaches, these organizations still have a way to go before their operations align with their AI strategy. Successful outcomes are correlated with adherence to operational best practices. Still, just one-third of respondents follow recommended procedures like machine learning operations (MLOps), redesigning workflows, and documenting AI model life cycles.

3. Orchestrate solutions for tech and talent

At this point in their journeys, organizations realize they need to leverage the resources they can get their hands on to keep up, whether human or technology resources.

Because most companies polled do not yet have a robust talent pool for AI, 53% prioritize hiring new outside talent with AI skills to aid their transformations rather than training existing employees to do those jobs. 65% of respondents reported that they acquired AI as an off-the-shelf product or contracted service rather than building their own artificial intelligence solutions.

In the long term, you will need to invest in developing AI talent in-house. However, for the time being, it will become increasingly challenging to hire the right talent easily. The best answer in the short run is to look outside for help. The three primary areas to focus on are AI platforms, AI products, and AI consulting companies (see How To Succeed With Enterprise AI: Buy Vs. Build.)

4. Select use cases for AI that accelerate outcomes

Finally, there is the matter of "which use cases to apply AI to first." The top AI use cases across all industries are: cloud pricing optimization (44%); voice assistants, chatbots, and conversational AI (41%); predictive maintenance (41%); and uptime/reliability optimization (41%). However, in choosing your use cases, what you want to look for are use cases that meet two critical criteria:

First, look for "low-hanging fruit." What applications of AI will be the easiest to implement or to be accepted by your organization? If you can get a win on the board early, it will be far easier to maintain support from above and below. Keep in mind the adage: garbage in, garbage out. Be sure to choose an area with enough data to train your AI models adequately.

Secondly, which applications of AI will deliver the most value for your organization? Demonstrating a strong return on investment for your efforts will create momentum for further investment and help foster an environment that is friendly to artificial intelligence.

If there's one thing in common to all the insights summarized here, it is that successful implementation and scaling of Artificial Intelligence initiatives relies as much or more on the human aspects of change as it does on the technological aspects. Like everything worthwhile, it won't come overnight. But the horizon for success here is shorter than you think. Whatever you do, "starting today" is the right decision.

If you care about how AI is determining the winners and losers in business, and how you can leverage AI for the benefit of your organization, I encourage you to stay tuned. I write (almost) exclusively about how senior executives, board members, and other business leaders can use AI effectively. You can read past articles and be notified of new ones by clicking the “follow” button here.

Follow me on LinkedInCheck out my website