Majority of businesses are still at early stages of AI maturity: MIT study

Although businesses have embraced AI due to its transformative abilities, the majority of them are still at the piloting stage for the technology, according to a recent MIT study.

The MIT study highlights that there are at least four stages in AI advancement by enterprises, and as these stages advance, so does the overall financial performance as well.

MIT says most companies are working through the initial stages

According to the MIT Center for Information System Research (CISR) study which analyzed 721 companies, most enterprises are still working through the first two stages of AI advancement. The study’s authors led by Peter Weill and Stephanie Woerner, both with MIT showed that these companies had financial performance below the industry average.

But businesses in the third and fourth stages of AI maturity however exhibited a financial performance above the industry average, exceeding 10 percentage points, according to the authors.

Of the surveyed companies, 28% were in the first stage, which was also categorized as experiment and prepare stage, where leaders look at addressing concerns such as ethics and skills to ensure a smooth path forward.

“In this stage, enterprises focus on educating their workforce, formulating AI policies, becoming more evidence-based, and experimenting with AI technologies to grow more comfortable with automated decision-making,” explained the researchers.

The study also revealed that 34% of the enterprises were in the second stage which is characterized by building pilots and capabilities.

At this stage, according to the study, enterprises start to “define important metrics, begin to simplify and automate business processes, and develop the enterprise capabilities they’ve learned.”

It is also at this stage that use cases are piloted, and work on leveraging enterprise data and developing APIs. According to the study, companies in this second stage averaged 2.2 percentage points below the industry average.

Foundation and SLMs start surfacing in the later stages

According to the study, at least 31% of the surveyed firms were in the third stage, which is where enterprises start to develop AI-driven ways of working. During the third stage, foundation models and small language models (SLMs) are introduced and applied to enterprise opportunities.

It is at this stage that AI becomes industrialized, that is, it becomes available and replicable across the business. According to the study, this stage involves work on building a core platform for AI, bearing in mind issues of transparency to decision-makers via dashboards, subsequently transforming the organizational culture to enhance innovative and data-driven thinking.

The study also found that firms in this third stage averaged 8.7 percentage points above the industry average.

According to the MIT team, the fourth and final stage is where companies become AI future-ready. This stage accounted for 7% of the total companies surveyed.

According to the report, this is the achievement stage where AI is embedded in “all decision-making throughout the enterprise.”

“They leverage proprietary AI internally, and many sell new business services based on this capability, the AI capability as a service, or both to other enterprises.”

MIT.

Businesses in this stage averaged 10.4 percentage points above the industry average. According to Forbes, navigating through these stages of a collaborative effort from across the enterprise as the technology can help optimize operations.

The authors of the report have cited some firms at the various stages of AI advancement such as DBS Bank, which has committed to carry out 1,000 AI experiments per year.

Land a High-Paying Web3 Job in 90 Days: The Ultimate Roadmap