How to use this: Use these MIT guides to assess your company’s AI maturity and decide which technologies to use and which skills to invest in.
Companies are changing their organizations to benefit from artificial intelligence, with 88% of respondents in a recent McKinsey survey saying their organization uses AI in at least one business function.
Whether your organization has already successfully implemented AI or is starting to consider using the technology, these three guides from MIT research can help you plan what’s next.
How mature is your company’s AI?
Researchers at the MIT Center for Information Systems Research developed this AI maturity model to help companies assess their AI capabilities and chart a path forward. Using a survey of 771 companies and interviews with executives at nine companies, researchers found that companies in the first two stages of AI maturity (experiment and prepare, pilot and build capabilities) had lower financial performance than the industry average. Companies that reached the latter two stages—developing AI ways of working and becoming ready for the future of AI—outperformed the industry average financially.
“We recommend gathering a team of senior technology and data leaders to assess which of the four stages your company is currently in, as well as your goals and timeframe for using AI,” says Stephanie Woerner, Principal Researcher at MIT CISR. “Then we discuss which company capabilities and skill sets need further work. It doesn’t matter where.” [with AI maturity]be bold. ”
To learn more about the four stages of AI maturity, see What is your company’s AI maturity level?
Generative AI or Machine Learning: Which AI Tools to Use and When?
Generative AI is widely accessible and useful. Even though ChatGPT has been publicly available for three years and agent AI is starting to grab headlines, large-scale chatbots built on large-scale language models still feel novel and exciting.
That doesn’t mean generative AI is a one-size-fits-all AI.
In a recent commentary on machine learning and generative AI, Associate Professor Swati Gupta and Professor of Practice Rama Ramakrishnan break down the differences between traditional machine learning and generative AI. We’ll discuss when each is most useful and how the two are best used together. We’ve compiled that guidance into a chart. It can be used to guide decisions about the use of AI in the early stages of project, initiative, or strategy development.
For more guidance from Mr. Gupta and Mr. Ramakrishnan, see Machine learning and generative AI: What are they good for?
Which human abilities best compensate for AI’s shortcomings?
Companies need to invest in upskilling their employees and creating roles that emphasize basic human abilities. MIT Sloan Postdoctoral Fellow Isabella Loaiza and MIT Sloan Professor Roberto Rigobon have grouped five human competencies that will become valuable as AI is implemented across industries, under the acronym EPOCH. An analysis of data from the U.S. Bureau of Labor Statistics found that jobs that relied on EPOCH capabilities had larger employment increases than jobs that did not.
To learn more about the EPOCH framework, see These human capabilities make up for AI’s shortcomings.
