iStockPhoto / Getty Images
New research from McKinsey shows that enterprise adoption of artificial intelligence — particularly generative AI tools such as ChatGPT — has taken a significant leap forward in the past year.
Sixty-five per cent of those surveyed reported that their companies regularly use some form of the technology, double the number who answered yes when asked the same question just 10 months earlier. But for all the accelerated adoption (and pressures to not be left behind), many business leaders remain nervous about AI’s role in the future of their organizations.
Potential enterprise AI benefits include increased efficiency, reduced ‘grunt work’ so employees can focus on higher-order activities, and enabling data processing at scale. Pitfalls are also a factor. How do you manage risk? How do you handle governance? How do you use AI as a tool to maximize human potential rather than replace it?
“While AI holds incredible potential, many business leaders feel overwhelmed by it,” says Jim Stratton, Chief Technology Officer at Workday. “AI can be complex and daunting, making it hard to know where to start or how to leverage it effectively.”
Mr. Stratton adds that Workday, which unites HR and finance on a single enterprise platform, has built AI into their technology for more than a decade.
“Workday Illuminate isn’t just an add on AI, it’s built right into the core of our platform. This means we can quickly deliver new AI capabilities that bring real value to our customers and to their businesses. And when it comes to data, our AI is trained on some of the industry’s largest and cleanest finance and HR datasets, helping to ground our solutions in reality. This approach offers distinct advantages when it comes to building AI systems, products and experiences.”
A wealth of experience means Workday is well-positioned to help organizations embrace the technology. It’s why the company launched its Workday AI Masterclass, a free resource that equips clients to become enterprise experts in AI, whether they’re new to the concept or seasoned practitioners.
“We understand the challenges that business leaders are facing when it comes to AI,” Mr. Stratton says. “Our goal is to demystify AI by offering practical insights into how it works and how it can be strategically integrated within your own operations.”
One key topic covered in the master class is how to choose the right large language models (LLMs) for an organization’s needs.
Large language models (LLM) are the underlying technology behind Generative AI that is designed for a variety of machine learning tasks. LLMs support tasks including search intent, topic classification, information summarization, and generative text.
“Enterprise data is unique because it represents how you and your organization run,” says Edward Raffaele, Workday’s vice-president of AI engineering. “It’s not shared with anybody else, and it contains the critical pieces of information needed to develop effective machine learning solutions that are tailored specifically to you.”
Implementing the correct LLM is essential to confidently solving problems and answering questions correctly.
“At Workday, we take a hybrid approach to implementing LLMs. That means we use partners to be able to bring in their best-of-breed models, partnerships like Google Cloud Platform and AWS. And then we also build our own large language models,” Mr. Raffaele says of the AI, called Workday Illuminate, that powers its HR and finance management platform. “By building our own large language models, we’re able to do a couple of key things. The first is to keep your data in-house. Second is to build smaller, more domain-specific models that are able to capture the nuance of your organization.”
A practical example he uses is writing job descriptions. “A model that we build internally is able to capture the nuance and tone of how you write your job descriptions, the syntax and structure of your job descriptions, and it’s also able to stay up to date as your job descriptions tend to update,” Mr. Raffaele says, adding it helps deliver on one of the promises of enterprise AI.
“Being able to not start from a blank page when you’re writing job descriptions or knowledge base articles, artificial intelligence can generate huge time savings. You can translate those into cost savings of running your business.”
Implementing AI in a responsible way is another common concern for organizations, and something Workday has put at the core of its own AI-powered platform.
“Responsible AI is an approach that guides the design and development of AI systems in a manner that adheres to transparent and accountable standards for trustworthy and ethical technologies,” says Kelly Trindel, Chief Responsible AI Officer at Workday. “The value of a proper a responsible AI framework is that it puts us on a path to innovate with integrity.”
Some pillars of this framework, according to Dr. Trindel, could include taking a risk-based approach to implementation, and ensuring implementation is transparent and clear.
“You want to build a risk identification and risk mitigation process that is robust, scalable, aligned with existing and developing regulations, tailored to your specific business and risk tolerance,” she says. “An effective responsible AI program actually speeds up innovation. It’s a false dichotomy that you can either be ethical or you can be fast. Responsible AI brings these two aspects together.”
It also needs to be a “human-centric” process, something underscored by Workday’s Aashna Kircher, Group General Manager, office of the CHRO.
“AI adoption is a team sport: up, down, and sideways. Everybody has to be on board in unlocking this strategy,” she says, pointing out that companies need to be ready for this technological leap forward. “Do you have cultural readiness? How do employees feel about AI? Do they understand the benefits or just have fears about this technology? Do you have a culture of experimentation and embracing new ways of working?”
Choosing the right partners and tools – such as Workday’s AI-assisted platform — can also help ease this transition.
“Look for partners who are building to solve enterprise challenges around data privacy, compliance and transparency,” Ms. Kircher says. “Understand the long-term philosophy and vision, as this space is evolving rapidly, and choose partners who are in it for the long game.”
Advertising feature produced by Globe Content Studio with Workday. The Globe’s editorial department was not involved.
link