Turning techy: Best practices for AI implementation
Developing, training teams and starting small are key
As artificial intelligence use becomes more common in the equipment finance industry, developing best practices is key to successful implementation.
AI can be used to simplify tasks for employees and improve operations without changing the way the business runs, Solifi’s Chief Technical Officer Eldon Richards said during a recent Innovation Roundtable on AI presented by the Equipment Leasing and Finance Association.
AI adoption is part of a larger industry shift toward innovation, Richards said.
“We need to be on the lookout for [AI evolution] when we’re thinking about innovations and how we keep our companies moving forward so that we are prepared for those shifts,” he said. “We can be the innovator, but also we can be aware that somebody else might be doing that. It’s somebody else now using autonomous cars to replace taxi drivers or Uber drivers, and that’s a revolutionary change.”
However, the launch of AI in the workplace for equipment financers requires developing best practices for successful implementation.
1. Develop a team
AI implementation requires in-depth knowledge of the organization and the technology. Due to complexities and the volume of information needed, having a dedicated team can make the initial implementation easier, Somerset Capital’s Chief Information Officer Andrew Cotter said during the Innovation Roundtable.
“Establishing a team that’s going to be pioneering this within your organization … I don’t think there should be just internal resources there,” he said. AI “is moving so fast, and there are experts that can follow this and it’s their job to do this.”
2. Education, education, education
AI education, especially as the tech continues to evolve and usage expands, is key to selection and implementation. AI can make almost any task more efficient, but the user and AI team need to understand what it is and how it can be applied, Solifi’s Richards said.
“Simply learning about the technologies, what they can do, learning about what others are doing and saying, ‘I can apply that, and I can make this process more efficient,’” is important, he said.
It’s also crucial to educate AI-driven technology users early on, lucy technologies founder and Chief Executive Harmony Oswald said during the Innovation Roundtable.
“People that are inputting the prompts really need to be somebody who’s educated about the industry and informed about AI in order to make the most efficient use of these tools,” she said.
3. Install controls
Organizations using AI must develop controls and guardrails for its implementation to ensure the technology is being used safely and ethically, Somerset Capital’s Cotter said.
“While there’s a lot of good that can be done [with AI], there’s a lot of bad that can be done and is being done,” he said.
As a result, organizations should vet AI software providers and review the terms of service, Oswald said. Checking issues such as IP address clauses, identification clauses and indemnification clauses can make a difference from a legal standpoint, she added.
4. Start small
When adopting AI, the first step is as simple as identifying a problem and finding a way for AI to solve it, Xiang Ji, commercial risk senior manager at Toyota Commercial Finance, said during the Innovation Roundtable.
“Starting with something small really helped [Toyota Commercial Finance] solve the problem,” she said. “We’ve successfully improved our [loan] approval rate by probably four times as compared to what we had before, without taking any more losses.”
