How to evangelize Artificial Intelligence (AI) in your organization
Blog: The Enterprise Project - Enterprise Technology
How to evangelize Artificial Intelligence (AI) in your organization
August 11, 2021 – 3:00am
Organizations seeing the most benefits from Artificial Intelligence (AI) work are more likely to be true believers in cognitive capabilities. Indeed, AI high performers, as identified by McKinsey, invested more of their digital budgets in AI than their counterparts, were more likely to increase their AI investments in the next three years, and employ more AI-related talent, such as data engineers, data architects, and translators, than their counterparts.
Winning over the end users of AI-enabled capabilities is just as – if not more – important to your success.
“Winning support for AI across the business is crucial for CIOs and other IT leaders hoping to scale their programs,” says Dan Simion, vice president of AI & Analytics at Capgemini North America.
How to make the case for AI and build support: 11 tips
To successfully earn buy-in and sponsorship from C-suite executives and line of business teammates alike, there are several moves IT leaders can make:
1. Create excitement within IT
The implementation of AI across the entire Lenovo organization is enabling greater efficiency and effectiveness. But, says Arthur Hu, Lenovo Group’s vice president and CIO, in the past, there has been a gap between strategy and execution. Keep your eye on the macro picture, he advises. “AI is constantly changing, so during strategic planning, I encourage my team to talk about the lifecycle of technology that is currently in use around the company,” says Hu, who asks his team to consider what’s ready for retirement and what’s headed for the mainstream. “By letting my team get creative, it naturally builds momentum and develops excitement.”
[ Check out our primer on 10 key artificial intelligence terms for IT and business leaders: Cheat sheet: AI glossary. ]
2. Get executives involved to match AI to digital transformation strategy
Success requires plenty of input from senior executives. “Executives need to work with their AI teams to ensure that the AI system’s input and output will be aligned with their overall digital transformation strategy,” says Justin Silver, AI strategist at AI platform provider PROS. “Collaborative strategy sessions that bring together executives and AI researchers can bring broader visibility to AI initiatives and keep AI research and development efforts tethered to key needs of the business.”
3. Assemble other advisory councils
“When making [changes] like this it is important to connect and host advisory councils with your organization’s partners, suppliers, and employees to gain their insight and opinions on implementation,” Hu says. “By doing so, you will see where it is wanted and needed.”
Understand that while IT can assist in AI-enabled innovation, IT cannot demand it of the business, says Hu. It must be co-created. “We have to remember to encourage applied curiosity, and then we can figure out our next steps for execution.”
[ Want best practices for AI workloads? Get the eBook: Top considerations for building a production-ready AI/ML environment. ]
4. Integrate AI teams into the organization
“Don’t put an AI team in a small silo and tell them to transform the business by themselves,” says Peter Scott, AI consultant and founding director of Next Wave Institute. “That’s like running power to one only one room and telling them to electrify the enterprise. AI is leveraging the intersection of technology and cognition, and it impacts every part of your business where people are thinking.”
5. Personalize the benefits
To encourage user adoption of AI throughout an organization, IT leaders and managers need to demonstrate how the resulting changes will benefit employees. Present quality data that shows how business processes can be improved. “Presenting a successful use case with support from senior leadership and easily recognizable data can incentivize usage,” says Chris Fielding, CIO at Sungard Availability Services. “IT leaders and managers should clearly communicate to team members why the incorporation of AI is beneficial and the positive impact it will have on productivity and efficiency both day-to-day and in the long run.”
Staying people-focused will go a long way, says Silver of PROS. “Retention of employees who have the skills and experience working with AI systems can yield positive returns.”
6. Address job loss concerns
“One of the challenges in AI adoption is the fear that many functional leaders have about losing their jobs or becoming obsolete,” says Nancy A. Shenker, author of Embrace the Machine. “When it’s positioned primarily as a technology improvement or cost-saving advancement, line managers who don’t fully understand the power of AI and ML will start to freak out and get defensive about the human elements of their jobs.”
By emphasizing that AI allows teams to focus on the actions to take based on the insights generated by AI and ML solutions rather than spending time mining the data for patterns, it becomes clear that AI does not remove the need for human decision-making, but instead enables a more effective, efficient route to insightful outcomes, says Simion of Capgemini.
Let’s explore five more tips:
7. Tout AI’s power to prevent human error
AI can significantly reduce human errors, which is critical in some industries wrestling with manual-based processes, such as healthcare. “Every time a human has to copy information from one place to another, there’s an opportunity for error,” says Karan Yaramada, CEO of Kanverse, a provider of cloud-based AI automation. “At best, the error is an administrative nuisance; at worst, it can have real consequences on the output.”
8. Focus on high-value use cases
“Your organization needs decision-makers who understand the technology, at least enough to know what is and is not possible,” says Silver of PROS. “Keep business leaders in the organization in the loop on AI initiatives to ensure that AI is being applied to high-priority, high-value use cases that are properly framed.”
Demonstrating desired business outcomes is key. “One way to win support for AI is to design a new operating model for the business that utilizes AI and ML solutions,” Simion says. “By leveraging AI and ML solutions, IT leaders can demonstrate the incremental value to business stakeholders who are heavily invested in [a] change in process.”
[ How can public data sets help? Read also: 6 misconceptions about AIOps, explained. ]
9. Build AI trust
Business leaders may be hesitant to take these outputs of AI or machine learning models seriously. “Instilling a sense of trust by working together with business leaders to demystify AI solutions, how they work, and how the outcomes are generated is key for changing the business mindset,” Simion says.
“When business leaders see the value of the outcomes being generated by AI solutions, then they can uncover the disruptive insights and be willing to implement them for decision-making that drives desired business outcomes to make a large impact across the organization.”
10. Emphasize real returns
The vast majority of organizations with AI or machine learning projects underway saw value from them in less than six months, according to ESG. “Decision-makers need to continuously assert the idea that implementing intelligent automated systems can help users become very productive by eliminating choke points in business processes,” says Yaramada of Kanverse.
11. Don’t overpromise; do share the quick wins
“Take the time to understand what’s realistic with AI and what isn’t,” says High. “AI can perform amazing feats in narrow contexts when you have lots of well-curated data.” Look for quick wins: use cases where detecting patterns or directional insights can still be actionable by the business user in use cases that previously could not scale, advises Kurt Trauth, senior vice president of customer experience strategy and analytics at Stratifyd, a developer of AI and analytics platforms. “Individuals will quickly recognize the potential ROI of integrating AI into their area of business,” Trauth says, “overcoming the often-incorrect assumptions that AI is too complex or costly to implement.”