Did you miss a session from MetaBeat 2022? Head over to the on-demand library for all of our featured sessions here.
Deloitte released the fifth edition of its State of AI in the Enterprise research report today, which surveyed more than 2,600 global executives on how businesses and industries are deploying and scaling artificial intelligence (AI) projects.
Most notably, the Deloitte report found that while AI continues to move tantalizingly closer to the core of the enterprise – 94% of business leaders agree that AI is critical to success over the next five years – for some, outcomes seem to be lagging.
For example, 79% percent of respondents reported achieving full-scale deployment for three or more types of AI applications, which is up from 62% last year. But the percentage of organizations in the “underachiever” category – those who have deployed a high number of AI projects but have low outcomes – rose from 17% last year to 22% this year.
That may seem to be a contradiction, but according to Beena Ammanath, executive director of the Global Deloitte AI Institute, it’s not a surprise. What is a surprise, she added, is how quickly the AI landscape is changing – to the point that what began as an every-other-year Deloitte report is now created annually.
Join today’s leading executives at the Low-Code/No-Code Summit virtually on November 9. Register for your free pass today.
“The AI space is evolving so fast and the technology is growing so rapidly,” she said. The fact that many companies are underachieving with the outcomes of their AI investments “tells you how fast the technology is being deployed.” The investment in AI is there, she explained, but there is so much happening in AI research and applications that many are unable to keep pace as far as gaining value.
Deloitte suggests 4 AI action steps
To fully fuel the AI transformation in the enterprise, Deloitte’s report focuses on four key actions that can power widespread value from AI, right away.
First, investing in culture and leadership is essential. “AI is not something that can be relegated to your IT or back-end teams,” said Ammanath. “This can have a real impact on your core business, it can create new revenue opportunities, new product ideas that just didn’t exist before.”
Instead, you need the business leadership to buy in, not just to get the technology into the company, but to drive the culture change and bring in the right talent.
“The entire organization needs to be on board with the technology, whether it’s the AI team deploying and building the tool or you have an AI team actually creating the models,” she said.
That hasn’t happened in most organizations yet. According to the report, while 43% of respondents report appointing a leader responsible for effective human and AI collaboration, concrete actions are lagging.
Transforming operations is key
According to the Deloitte report, high-outcome organizations are “significantly more likely to adopt additional operational leading practices.”
These include tracking the ROI of deployed models and applications. Eighty-six percent of high-outcome enterprises do, compared to only 71% of low-outcome organizations.
High-outcome organizations are also more likely to have a documented process for governance and quality of data put into AI models, follow documented MLops procedures and a documented AI model life cycle publication strategy, leverage a common and consistent platform for AI model and application development, and use an AI quality and risk management process and framework to assess AI model bias and other risks before models go into production.
The results are especially relevant, the report pointed out, given that a clear majority (60%) of respondents viewed AI solutions as strategically “very important” for their organizations’ success, including more than 55% of respondents from low-outcome organizations.
Orchestrating tech and talent together
The Deloitte report found that organizations need to plan AI technology and talent investments in tandem, understanding that both human talent and AI-driven technology have specific skill sets.
The ability for an organization to achieve differentiated tools and applications with AI still hinges in large part on the talent it is able to bring in-house, the report said, but Ammanath also emphasized the importance of making sure that everyone in the organization understands basic AI.
“You [may not] need to know what a diffusion model is, for example, but you need to have a high-level understanding of what AI is … that basic foundational training should be something that you provide to every employee,” she said. Then, organizations need to provide role-based training, so that a data scientist that supports a marketing team, for example, understands the specific questions needed to evaluate a vendor.
Finally, organizations need to provide governance mechanisms, so employees aren’t expected to know everything about every kind of AI tool. “There are tactical things you can do to actually start addressing these issues and make this less complicated for your employees right now,” said Ammanath.
Select the right AI use cases
Finally, the Deloitte report cautioned that as companies increase their AI investments, they eventually address a larger and wider range of use cases. Organizations should be choosy about selecting which business processes to start with, since the choices they make now may set the course for how quickly they will achieve successful outcomes and gain momentum.
“Doing AI for the sake of AI is never a good idea,” said Ammanath. “You know, you start buying some of these tools or setting up a Data Science Center of Excellence without having a clear pathway towards the value – whether it’s not deploying it in the right function, or not putting the right level of thought into it – it can actually create a false start and impact the end goal.”
Deloitte’s Ammanath remains optimistic
Ammanath says she is an “AI optimist” who sees a lot of potential among organizations that are currently underachieving in their AI project outcomes.
“I do hope that underachievers are able to tap into the ecosystem, learn from these best practices and get more value,” she said. “I honestly believe that AI can bring a lot of value, and we can address the risks, but we also need to focus on the value creation.”