The adoption of artificial intelligence (AI) has more than doubled since 2017. However, the number of companies using AI has stayed about the same for a few years reveals a study by McKinsey. The companies that are doing well with AI continue to do better than their competitors. This is because they are making larger investments in AI, engaging in more advanced practices, and hiring more people with expertise in AI.
AI impact and investment during the last five years
AI use cases and capabilities
The number of AI capabilities that organizations are using, such as natural-language generation and computer vision, has doubled in the past four years. The most commonly used capabilities are robotic process automation and computer vision. Natural-language text understanding is also becoming more popular.
However, the top use cases have remained relatively stable. The top spot has been taken by the optimization of service operations for four years in a row.
AI spending
The amount of money that organizations spend on AI has increased as its use has become more common. Five years ago, 40% of people who used AI said that more than 5% of their digital budget was spent on AI. Now, more than half of the respondents say the same thing. 63% of people say that their organization’s investment in AI will increase over the next three years.
Value from AI
In 2018, manufacturing and risk were the two functions in which the largest number of respondents reported seeing value from AI use. Today, marketing and sales, product and service development, and strategy and corporate finance report the biggest revenue effects. Respondents report the highest cost benefits from AI in supply chain management. The value realized from AI remains strong overall.
AI-related risks
Despite the increased use of AI, there has been little change in the amount of risk mitigation that organizations are doing to make sure that their digital trust is secure.
The gap between AI leaders and laggards continues to grow
High-performing companies are well-positioned for sustained success with AI. They can develop AI more efficiently and create a more attractive environment for talent. By following them, other organizations have a clear blueprint of best practices for success.
- AI leaders are expanding their competitive advantage and are more likely to follow core practices that help them unlock value. This includes linking their AI strategy to business outcomes. They also engage in “frontier” practices more often which enables them to develop and deploy AI at scale.
- High performers are automating most data-related processes to improve efficiency in AI development. They are also expanding the number of applications they can develop by using more high-quality data to feed into AI algorithms.
- AI high performers engage non-technical employees by 1.6 times more than other organizations to create AI applications using emerging low-code or no-code programs, which accelerate the creation of AI applications.
- High performers might also be capable of managing potential AI-related risks, such as personal privacy and equity, and fairness. Other organizations have not addressed these risks yet.
- Investment is another area that could contribute to the widening of the gap. AI high performers are expected to spend more than other organizations on AI efforts. They are nearly eight times more likely than their peers to spend at least 20% of their digital-technology budgets on AI-related technologies. They are five times more likely than other respondents to spend more than 20% of their enterprise-wide revenue on digital technologies.
Suggested reading: IDC finds an emerging divide in organizations’ AI capabilities, separating leaders from laggards
What organizations are doing to source AI-talent
Software engineers (39%) are the role that organizations have hired most often in the past year than data engineers (35%) or AI data scientists (33%) when it comes to AI. This is a clear sign that many organizations have shifted from experimenting with AI to actively using it in enterprise applications.
There is a shortage of talent in the technology industry. This is making it harder for some companies to make the shift to using new technology. Most organizations say it has been hard to find workers for every role related to AI in the past year. The hardest role to fill is the data scientist, according to most people surveyed.
Upskilling and reskilling employees
Both AI high performers and other organizations are upskilling and reskilling technical and nontechnical employees on AI as a way of gaining more AI talent. However, high performers are taking more steps than others to build employees’ AI-related skills. They are nearly three times more likely than others to have capability-building programs to develop technology personnel’s AI skills. The most common approaches include experiential learning, self-directed online courses, and certification programs. Other organizations most often depend on self-directed online courses.
Building a diverse AI team
There is a scope for significant improvement when it comes to the level of diversity within organizations’ AI-focused teams. According to the survey, only 27% of employees working on AI solutions are women, and just 25% are from a minority group. Shockingly, 29% of respondents said their organization had no minority employees working on their AI solutions.
Some companies are working to improve the diversity of their AI talent. 46% of respondents say their organizations have active programs to improve gender diversity in AI teams. They are partnering with diversity-focused professional associations to recruit candidates. One-third of organizations have plans of increasing racial and ethnic diversity.
AI high performers are 3.5 times more likely to have at least 25% of AI development employees as women. They are also twice more likely than other organizations to have 25% of AI development employees as racial or ethnic minorities.
Source: McKinsey and Company
Also read: Google lists 3 ways by which AI can make technology more accessible and useful
