Gen AI’s next inflection point: From employee experimentation to organizational transformation

After nearly two years of debate, the verdict is in: generative AI (gen AI) is here to stay, and its business potential is massive. We’ve already witnessed an exponential rate of gen-AI-related innovation, which promises to accelerate automation and enhance productivity, innovation, and the quality of work, as well as the employee and customer experience. The companies that fail to act and adapt now will likely struggle to catch up in the future.

Despite all the buzz, most companies have yet to scratch the surface of gen AI’s promise. A recent McKinsey Global Survey reveals that employees are far ahead of their organizations in using gen AI,¹The online survey was in the field from February 27 to March 8, 2024, and garnered responses from 592 participants representing the full range of regions, industries, company sizes, functional specialties, and tenures. Of those respondents, 127 say they are using publicly available or internal gen AI tools almost always at work; 51 say they use public tools never or rarely and that they never use internal tools, or that internal tools are not available to them; and the other 414 say they use either internal or public tools sometimes, often, or at varying frequencies by the type of tool. as companies have been slow to adopt in ways that could realize gen AI’s trillion-dollar opportunity. To harness employees’ enthusiasm and stay ahead, companies need a holistic approach to transforming how the whole organization works with gen AI; the technology alone won’t create value. This means applying gen AI in ways that enable the business strategy: by reinventing operating models and entire domains,²That is, specific workflows, processes, journeys, and even functions. by reimagining talent and skilling, and by reinforcing changes through robust governance and infrastructure.

Employee use is at an inflection point, while their organizations lag behind

According to our research, employees are forging ahead with gen AI, a broadly accessible technology that puts AI’s potential at everyone’s fingertips. Nearly all respondents (91 percent) say they use gen AI for work and the vast majority are enthusiastic about it (Exhibit 1). Nine in ten also believe the tools could positively impact their work experience and most believe gen AI will help with a range of skills, from critical thinking to creativity.

1

In this respect, most companies are lagging behind their employees. As high as employee usage is, organizational maturity with gen AI is strikingly low. In our survey, only 13 percent of respondents’ companies have implemented multiple use cases, a group we call “early adopters” (Exhibit 2).³We define early adopters as those companies, according to respondents, that have implemented six or more gen AI use cases to date. Among them, there’s a larger share of heavy users: that is, employees who use either public or internal gen AI tools every day or two. Compared with others, this group is likelier to use gen AI for a range of work activities and report greater productivity gains.⁴Survey respondents were asked to rate the frequency of their use of gen AI tools, both publicly available and internally developed tools, at work—as well as their use of public tools for nonwork purposes. For all of these questions, potential responses were “never,” “rarely (that is, once per month),” “sometimes (that is, once per week),” “often (that is, two to three times per week),” and “almost always (that is, every day).”

2

The chief information officer of a global heavy industry company sees these trends at his own organization. Employees are experimenting with gen AI through publicly available and embedded tools,⁵Such as OpenAI’s ChatGPT and Microsoft’s Copilot. which is increasing curiosity and encouraging greater openness to experimentation. Yet he notes that there’s no easy-to-prove business case for employee-driven adoption and the piecemeal implementation of use cases.

The next inflection point: Moving from individual experimentation to strategic value capture

Technology adoption for its own sake has never created value, which is also true with gen AI. Whether technology is itself the core strategy (for example, developing gen-AI-based products) or supports other business strategies, its deployment should link to value creation opportunities and measurable outcomes (for more, see sidebar “‘People led, tech powered’: Walmart’s vision for gen AI”). Our survey findings suggest that early adopters are on track: 63 percent of early-adopter respondents say their organizations’ AI and gen AI strategies align greatly with their business strategies, compared with only 17 percent of respondents at “experimenter” companies.⁶We define experimenters as those companies, according to respondents, that have implemented one to five gen AI use cases.

To capture gen AI’s full potential, companies must consider how the technology can redefine the way the organization works. Our experience and research point to three steps to prepare for gen AI’s next inflection point: reinvent the operating model by translating vision into value, domain by domain; reimagine the talent and skilling strategy; and reinforce changes through formal and informal mechanisms that ensure continuous adaptation.

Reinvent domains by translating vision into value

Companies can only reap gen AI’s full benefits, which range from faster innovation and enhanced productivity to improved employee and customer experience, when they use technology to make transformative changes. More specifically, this means embracing holistic changes to the operating model, including key processes, ways of working, capabilities, and culture. Because anyone can use gen AI, these tools can act as a gateway technology for all other digital and tech transformations.

To start, companies should prioritize the right unit of transformation by focusing on specific domains, such as product development, marketing, and customer service. This domain-based approach allows for end-to-end, technology-led transformation that integrates multiple use cases within a single value-creating workflow, process, journey, or, occasionally, entire function. Since domains often span organizational boundaries, implementing gen AI and other technologies at the domain level can deliver greater value than one-off solutions.

Here are examples of what’s possible with a domain-based transformation, and the implications for roles and day-to-day work:

• In software development, gen AI can revolutionize work by delivering higher-quality, resilient products much faster; think days instead of months. This will require changes across the product life cycle and closer collaboration between product and engineering teams. Comprehensive product data, prompt-based proofs of concept, and automated requirements can shorten ideation-to-prototyping timelines, allowing for a greater number of iterations. The use of self-writing code, autogenerated user guidance, and continuous code testing would also transform engineers from task completers to systems designers.
• In marketing, gen AI could (finally) enable the vision of true personalization at scale. Companies such as Netflix and Spotify have started on this path with hyperpersonalized video previews and personalized user playlists. These types of practices can enhance engagement and loyalty, allow brands to integrate seamlessly into customers’ lives, boost productivity of content creation, and improve ROI across the sales and marketing funnel. By doing so, the marketing organization’s silos could break down, especially between the creative and analytics teams.
• In customer service, gen AI can transform teams into centers of customer delight by proactively addressing issues and offering new products, all at reduced costs. AI-empowered humans will work with gen AI agents, using real-time trends and customer insights to become empathetic problem solvers and supervisors of customer experience. In the process, customer service agents and supervisors will make more use of technology, apply systems thinking, balance empathetic and commercial mindsets, and work more closely with the customer experience and product teams.
• Gen AI is also set to revolutionize cross-cutting domains, such as performance management and team management. For the latter, gen AI can put coaching prompts at managers’ fingertips and make it easier to access employee resources. This can meaningfully shift the time managers spend on certain tasks: less on administrative to-dos, more on checking in with team members and developing soft skills.

Reimagine talent and skilling by putting people at the center

As the examples above highlight, gen AI’s implications for talent and skill needs are massive. The technology’s potential to accelerate automation and transform operating models will significantly affect the roles and skills that organizations need. According to other McKinsey research, half of today’s work activities could be automated between 2030 and 2060, accelerating previous, pre-gen-AI projections by a decade. This puts pressure on organizations to understand their talent and skill needs quickly, adopt various strategies to close skill gaps, and invest in upskilling and reskilling. A gen-AI-based talent transformation isn’t something companies can simply hire their way out of, as it affects the entire organization and its ways of working.

Our research shows that early adopters prioritize talent and the human side of gen AI more than other companies (Exhibit 3). Our survey shows that nearly two-thirds of them have a clear view of their talent gaps and a strategy to close them, compared with just 25 percent of the experimenters. Early adopters focus heavily on upskilling and reskilling as a critical part of their talent strategies, as hiring alone isn’t enough to close gaps and outsourcing can hinder strategic-skills development. Finally, 40 percent of early-adopter respondents say their organizations provide extensive support to encourage employee adoption, versus 9 percent of experimenter respondents.

3

Companies can capitalize on employees’ enthusiasm for gen AI by investing in both technology adoption and skills (for more, see sidebar “Taking the granular view on gen AI’s workforce implications”). As previous McKinsey research shows, macroeconomic investments in both enable productivity gains that organizations can also see. This will require a tailored approach to reskilling and upskilling and close collaboration between business and tech leaders and HR. Given the criticality of people topics, HR plays an especially important role in gen AI and technology transformations, both by transforming the people domain and by acting as a gen AI copilot for all employees. One executive noted that for every $1 spent on technology, $5 should be spent on people.

With gen AI, building capabilities across the entire enterprise is crucial. As it’s a rapidly evolving, widely accessible technology, employees must adapt to the new skills (such as prompt writing, contextualization, and data-driven decision making) that gen AI demands. While specific skills shifts will vary greatly by company, all organizations will need to take a dynamic approach to talent development, based on their operating-model transformations; building skills is an ongoing process. As gen AI and automation reshape roles, employees will also need strong cognitive, strategic thinking and social and emotional skills to handle more complex tasks that complement AI.

Within specific roles, the tech talent who are scaling gen AI and future technologies will need to build, train, and fine-tune AI models. These newer skills will require immersive learning in areas such as software development, cloud integration, and security. Tech talent must also be able to contextualize and apply their judgment when translating business needs into technology solutions. Furthermore, companies will need tech-adjacent roles to manage the governance, operational, HR, and legal aspects of AI. Some roles, such as chief AI officers, will be brand new.

For domain-based talent, many will need intensive upskilling as their roles evolve. This will include different types of on-the-job learning and formal training opportunities. For example, healthcare professionals might take courses on personalized treatment planning and AI-driven diagnostics that are supplemented with mentoring and real-world projects.

And for all employees, including leaders and managers, it’s vital that everyone learns to use gen AI effectively and safely. Examples include comprehensive learning programs that cover responsible use and effective interaction with AI, as well as more augmentation-focused trainings, such as using gen AI coaching that allows managers to practice giving feedback.

A European telecommunications company put tailor-made reskilling into practice by implementing an AI coach for its customer service agents. By analyzing call transcripts from frontline employees, the AI coach assessed people across 20 different soft and hard skills. Both team members and leaders could access a dashboard that tracked progress on these skills and delivered real-time feedback using customer quotes and examples. The AI coach also suggested improvements and learning content based on agents’ performance and behavior, creating a hyperpersonalized learning experience. This tool resulted in a 10 percent reduction in average handling time, a 20 percent increase in customer satisfaction, and a 15 percent increase in the rate of first-time-right responses.

Reinforce the changes to continue transforming

How, exactly, should organizations tackle these massive transformational changes? Real success with gen AI requires a comprehensive, integrated approach to creating value. Our survey indicates that the most useful enabler of future adoption is better integration of gen AI into existing systems, cited by 60 percent of respondents. To make gen AI changes stick, organizations need the right infrastructure to support continuous change and win over hearts and minds.

McKinsey’s original survey research

Read more

The first step is establishing the right governance for gen AI (for more, see sidebar “Good gen AI governance at work”). In our experience, this means creating a centralized structure that oversees the organization’s AI adoption, sometimes with a chief AI officer leading these efforts. Nearly all early-adopter respondents (91 percent) say they have implemented some governance structure for gen AI, compared with a smaller share (77 percent) of experimenters. A centralized model with a gen-AI-dedicated center of excellence helps align AI vision with execution. This model also facilitates the implementation of strategy, continuous measurement, adaptation to new insights, and further experimentation—specifically, which experiments to scale or to stop, based on priorities and risks.

The second step is treating these changes like a true transformation. This means defining the transformation’s infrastructure, roles, and measurement criteria; ensuring accountability within business units; and implementing a regular cadence to monitor progress—and adjusting as needed.

Third is addressing employee mindsets and behaviors across the organization. We know from extensive transformation research and countless conversations with executives that changing mindsets and behaviors is vital to any successful transformation. Indeed, in our survey, early adopters focus more than others on the four tenets of the influence model that enables such changes: role modeling, fostering understanding and conviction, building capabilities, and reinforcing new ways of working (Exhibit 4).

4

In the gen AI context, this means:

• Role modeling. Leaders should visibly adopt generative AI in their own ways of working. For instance, using AI tools to generate insights and make data-driven decisions showcases the technology’s benefits. It sets a strong example when a CEO uses AI to streamline workflows or a senior executive uses AI-driven analytics for business reviews, encouraging others to follow suit.
• Fostering understanding and conviction. Organizations should communicate the reasons behind implementing gen-AI-related changes through internal communications, town hall meetings, and training sessions. Highlighting AI’s potential to improve efficiency, accuracy, and decision making aligns the team with the new direction. Informative content such as video tutorials and success stories can build collective conviction in AI’s advantages.
• Building capabilities. Successful AI adoption requires comprehensive training programs. This includes training on data analysis, machine learning algorithms, and understanding AI-generated outputs. Collaborating with online education platforms to provide courses and setting up internal AI boot camps for hands-on experience ensures proficiency in AI technologies.
• Reinforcing new ways of working. Companies should integrate AI goals into performance metrics and evaluation processes. They can set targets related to AI adoption, measure AI’s impact on key performance indicators, and recognize employees who effectively incorporate AI into their work. For instance, sales teams could set targets that include leveraging AI for customer segmentation and lead generation, with bonuses tied to successful AI-driven strategies. Tracking and celebrating milestones such as efficiency gains or innovative AI applications embeds these practices within the organization’s fabric.

No matter where an organization is on its gen AI journey, the time for making transformational change is now. Employees are already asking their organizations for more, and some companies have begun moving from experimentation to value capture. By gen AI’s next inflection point, the downside of lagging behind—and missing out on gen AI’s potential benefits—may be even greater. With employees’ embrace of gen AI and the technology’s rapid evolution, companies can capitalize on the current momentum by addressing organizational barriers to adoption, which requires no less than fundamentally transforming the company’s operations and preparing people for continuous change.