Meet the inaugural class of the Frontier Firm AI Initiative

To set the curriculum, we’re announcing the Digital Data Design Institute at Harvard Frontier Firm AI Initiative, hosted at Harvard Business School in collaboration with Microsoft. The Initiative will research human-AI collaboration, upskill global C-suite leadership, and deliver new insights and tools to help reinvent organizations as Frontier Firms. 

“Executives that go all in on AI without a clear path forward risk falling into a frustrating cycle of pilots that don’t deliver value and have no impact,” says Karim Lakhani, Founding Chair of the Digital Data Design Institute at Harvard. “With this Initiative, we aim to create rigorous, evidence-based blueprints for high-performing human-AI workplaces, bridging the gap between ambition and true competitive advantage.” 

As Jared Spataro, Microsoft’s Chief Marketing Officer of AI at Work, puts it: “It’s no longer a question of ‘if’ AI is right for business—leaders today are grappling with ‘how’ to become a Frontier Firm.” 

The 14 companies here—the Initiative’s inaugural class—are reshaping how work gets done through three emerging patterns: empowering every employee with an AI assistant, amplifying impact with human-agent teamwork, and reinventing business processes with agents. Together, we’ll experiment, learn, and chart the course ahead.  

Snapshot: The 330-year-old global banking powerhouse is enabling seamless collaboration between humans and agents to enhance customer service, empower its employees, and drive innovation.  

Why they enrolled: By joining forces with other Frontier Firms, Barclays hopes to compare notes and progress faster as a group than anyone could alone—collaborating on tough challenges, discovering new opportunities, and influencing industry standards. 

Frontier focus: Barclays’ goal is to make banking more personal and intuitive by combining the best of digital convenience and genuine human engagement. “We want to make our engagement with customers and clients as easy, intuitive, and personalized as possible,” says Craig Bright, one of Barclays’ Co-Chief Operating Officers. “Being able to do that in their language, their dialect, at the time of day that suits them is important, and that’s what AI allows us to do,” adds Anne Marie Darling, Barclays’ other Co-Chief Operating Officer. “But so, too, is recognizing the point in that conversation where we need to shift them to a human.”  

AI bona fides: With almost 100,000 employees already using Microsoft 365 Copilot, Barclays is embedding AI into everyday workflows to help teams across functions work more efficiently. They’ve rolled out AI assistants in contact centers that summarize customer calls and guide human agents in real time, suggesting answers or next best actions. On the tech operations side, the bank used a squad of agents working with human experts to migrate from a legacy database system to a more modern platform in just 20 days—a task that used to take three months.  

Snapshot: One of the world’s largest banks is using AI to build a more secure, more efficient banking infrastructure. 

Why they enrolled: To collaborate with innovative industry leaders and offer its executives bespoke learning opportunities.  

Frontier focus: BNY’s big ambition is to run a highly efficient, AI-first bank. They hope to support more clients at speed and scale so transactions flow seamlessly and every process runs as efficiently as possible. Once they do that, says Chief AI and Data Officer Sarthak Pattanaik, “we plan to ultimately rethink entire products and eventually entire businesses as AI-first.” 

AI bona fides: BNY has already put digital labor on the org chart, with 150 “digital employees” working across the bank. These autonomous agents—who report to human managers and have unique employee IDs—collaborate to complete a workflow, initiating work and reacting to real-time client needs and system triggers. One digital engineer continuously scans BNY’s code base to find opportunities for code optimization and security reinforcement and can even write and implement improvements for low-complexity issues.  

Snapshot: The health insurer is committed to providing healthcare coverage that’s simple, reliable, personalized, and truly improves health outcomes. 

Why they enrolled: To lead in Frontier healthcare and learn from peers, share perspectives, and collaborate with the Initiative on research that accelerates ethical AI adoption—not for AI’s sake, but to make medical, pharmacy, and behavioral care more affordable and accessible for its customers. 
 
Frontier focus: For Cigna, AI success means scaling use cases that drive measurable impact to deliver on its mission. One way to do that: use AI and agents to simplify processes so employees can focus on providing personalized customer experiences. Think: agents that can help with appointment scheduling, medication refills, and reminders, while advocates help with coordinating complex medical needs and care. 
 
AI bona fides: Two-thirds of eligible members use Cigna Healthcare’s virtual assistant in the myCigna portal, and over 80% find it helpful. Other initiatives include synthesizing extensive plan information so advocates can more quickly answer customer questions and using AI insights to personalize care across medical, pharmacy, and behavioral services. The company is also exploring and leveraging predictive models to anticipate member needs. 

Snapshot: The global law firm is reimagining the legal profession and creating a culture that empowers the first Frontier lawyers.  

Why they enrolled: To exchange notes with their peers and shine a light on blind spots in their AI strategies and thinking. “By understanding what works well—or what doesn’t—we can accelerate our strategy to deliver legal services, which create distinct advantage for our clients,” says Paul Greenwood, Clifford Chance’s Chief Technology Officer.  

Frontier focus: “The lawyer of the future” at Clifford Chance will work in tandem with AI—leveraging agents to handle research and first-draft prep, allowing lawyers to concentrate on higher-value strategic counsel and critical advisory work. Ultimately, the firm aspires to create real competitive advantage for its clients and to use AI insights to help them see what’s around the corner. 

Clifford Chance is also working with the Initiative to measure Copilot productivity and efficiency while upholding the firm’s AI code and principles.

AI bona fides: Clifford Chance has already adopted Copilot at scale, with 90% of employees actively using it today. Across the firm, lawyers have formed grassroots “innovation groups” and built their own prototype agents to automate some tasks. They’ve also used AI to redesign existing services: a regulatory compliance agent now automatically tracks changing laws and generates tailored impact assessments, reducing the time spent by trainees compiling 300-page PowerPoints. What was once a labor-intensive, unbillable task is now a popular subscription offering for clients. 

Snapshot: DuPont is transforming its 200-year legacy to accelerate growth—embedding AI into everyday work to help deliver cleaner water, stronger infrastructure, and better healthcare. 

Why they enrolled: To accelerate DuPont’s shift from manual workflows to AI-powered operations, boosting performance without waiting for multi-year system overhauls.  

Frontier focus: AI is a strategic lever for what the company refers to as the “New DuPont,” directly supporting its three-year growth goals—margin expansion, faster innovation cycles, and deeper customer value. “Our commitment extends beyond technology,” says CIO Matt Abbott. “We’re upskilling our workforce to embed AI into everyday work so we can deliver with more consistency, agility, and impact.” 

AI bona fides: DuPont is already seeing big AI wins. Predictive maintenance is strengthening uptime and reliability across key facilities. In R&D, AI is enabling scientists to solve complex challenges faster, accelerating time from ideation to experimentation. Next, DuPont is deploying a new generation of functional agents across sales, customer service, procurement, IT, and finance to increase speed, precision, and customer value.

Snapshot: Eaton, which manufactures technology that keeps power flowing and machinery running, is using AI to streamline engineering and operations.

Why they enrolled: Eaton wants to establish itself as the premier intelligent power management company, using AI to manage and optimize how energy is generated, stored, and used. Says Katrina Redmond, Eaton’s CIO: “We want to learn and glean as much as we possibly can to enable Eaton on this journey.” 

Frontier focus: Eaton plans to save millions of dollars with AI in the next 12 months by decoupling their growth objectives from their cost line. “We can radically change that by using AI to help with our growth,” Redmond says. The firm projects that by 2030, 80% of transactions with its partners will be completely agentic, with agents talking to agents. 

AI bona fides: Eaton has already achieved impressive productivity gains and begun to reinvent workflows with AI and agents. Across its tens of thousands of employees (including 14,000 with Copilot access), Eaton has embedded AI into product design, manufacturing, sales, marketing, and support. So far, Copilot and agents have cut product design time by more than 40% and reduced customer support response times by 20%. And a custom “tariff AI” agent scans political news and trade rules to recommend shifting production among Eaton’s more than 200 factories, saving millions in costs.  

Snapshot: Pharma giant Lilly is building a digital workforce powered by AI to turn science into medicine for people around the world.  

Why they enrolled: “If you look back throughout our 150-year history, we’ve long used technology to scale,” says Diogo Rau, Lilly’s Chief Information and Digital Officer. “We scaled the Salk polio vaccine globally and the world’s first commercially available insulin.” Joining the Initiative was a chance to continue that legacy by shaping the future of healthcare innovation with AI. 

Frontier focus: Use AI to bring medicines to patients faster. That means speeding up every step of the process, from identifying promising molecules to conducting clinical trials to scaling up manufacturing. “Our scientists may be able to identify medicines that a human couldn’t find on their own because there's just not enough time,” says Rau. 

AI bona fides: More than 40,000 people at Lilly now work with Copilot, alongside more than 150 agents supporting research, trials, and operations. The company is also advancing agent-to-agent processes, where AI systems automate workflows while humans oversee and guide outcomes. Lilly is also building the most powerful AI supercomputer in pharma. 

Snapshot: Global professional services organization EY has made AI a board-level priority, pairing multibillion dollar investments with organization-wide adoption. 

Frontier focus: Use agentic AI to help humans direct teams of agents that run complex business processes—unlocking new ways to transform their business so they can help clients transform theirs. 

AI bona fides: EY deployed Copilot to more than 150,000 of its employees and is expanding rapidly worldwide. AI now accelerates workflows, from tax and legal research to deal management, HR and document drafting, driving a 15% productivity boost—the equivalent of thousands of manual hours. That time is being reinvested into client delivery, learning, and operational excellence. 

Snapshot: GHD is a global professional services firm in engineering, architecture, environmental, and construction solutions, operating across five continents. 

Why they enrolled: To stay ahead of exponential change while holding fast to its tradition of technical excellence. 

Frontier focus: Harness AI to meet global demand for sustainable energy, water, and communities—delivering verifiable results clients and the public can trust. 

AI bona fides: With 5,000 Copilot seats and active usage above 90%, AI is quickly becoming part of daily work at GHD. Teams are piloting agents for coaching, information retrieval, and drawing reviews—including critical checks like comparing draft versions and confirming technical work products include essential information. 

Snapshot: The global market research firm helps brands grow by translating consumer data into insights and marketing strategy. 

Why they enrolled: To hear from organizations leading in AI, compare notes, and create a deeper understanding of human behavior. “You can’t do it on your own,” says CTO Mark Kimber. “By the time you’ve spent 12 months learning, the market has moved 24 months beyond you.” 

Frontier focus: Kantar aims to reinvent how marketing data turns into insight, transforming what was once a slow, manual process into a driver of faster, smarter growth. 

AI bona fides: Building on years of machine learning experience and advanced analytics work, Kantar gave Microsoft 365 Copilot to all of its 20,000 employees in the last year. The result? Faster research, smarter insights, and unexpected creativity. Tasks like data cleaning, initial analysis, and report drafting can be done in a fraction of the time, allowing experts to spend more time on interpreting insights and advising clients. The cultural shift to an “AI-first” workforce is already delivering sharper insights to brands. 

Snapshot: Having shaped fashion and culture since the Gold Rush, the denim icon is out to write the next chapter: turning a legacy brand into an agile, insights-driven, direct-to-consumer retailer.  

Frontier focus: Use AI to level up from $6.5 billion to $10 billion in annual revenue by tapping their nearly 175-year-old well of brand data. 

AI bona fides: Levi’s didn’t wait for a top-down rollout—they handed Copilot to their employees and let them run. Agents are being built and deployed across the business, enhancing onboarding guides, helping retail employees better serve consumers, and more. One UX designer built an agent in Copilot Studio to discover and summarize hundreds of pages of consumer and industry research and answer questions on demand. Levi’s is now building a super-agent: Employees ask their question in one place and it connects with the right agents and returns the answer. 

Snapshot: Lumen’s secure, high-speed network moves data, powers apps, and makes AI work at scale for enterprises. 

Frontier focus: Lumen’s approach to AI is all about speed, scale, and empowering employees to do more strategic work. They rapidly experiment with bold ideas, learn fast, and turn pilots into enterprise-wide transformation. “Evidence beats anecdotes,” says Chief Revenue Officer Ashley Haynes-Gaspar. “Run controlled trials for each agent use case. Define outcomes, segments, guardrails, and decision rules. Scale winners—and stop what doesn’t work.” 

Their big ambition? Unite the power of world-class talent with the intelligence of AI to fuel growth and deliver better results for their customers. 

AI bona fides: In late 2023, the telecommunications company rolled out Copilot to thousands of employees in a single sweep. The next year, they made GitHub Copilot available to all engineers, reducing the cycle time of projects by an average of 20-30%. Across the company, employees have completed more than 35,000 AI trainings. Lumen is now partnering with the Initiative on an agentic AI prototype ecosystem. Custom-built AI tools will automate customer renewals, reduce onboarding time for new sales engineers, and streamline knowledge transfer, democratizing expertise across the organization.  

Snapshot: The payments technology company has made agentic commerce central to its strategy, giving customers intelligent shopping experiences powered by AI.  

Why they enrolled: To accelerate responsible AI adoption, shape industry best practices, collaborate with leading academics and peers, share insights, and influence the future of human-AI collaboration.  

Frontier focus: Embedding secure, agentic payments inside everyday interactions so people can shop, book and buy within Copilot and other AI platforms—allowing humans to lead and agents to operate. 

AI bona fides: Mastercard has spent the past year developing practical AI use cases, with more than 18,000 employees using Copilot to boost productivity and reinvent core processes from fraud prevention to product development. Its new Agent Pay feature moves payments inside the conversation, allowing agents to conduct transactions on behalf of users. Product, partnerships, and data teams set the standards, while agents help customers curate options, verify credentials, and ensure secure checkout. The result: a blueprint for intelligent transactions that keeps people in charge and agents on task. 

Snapshot: The global food and beverage leader is using AI to make its operations—from how food is grown to how it reaches consumers—more efficient, sustainable, and connected.   

Why they enrolled: “No company can do AI alone,” says Chris Wright, Nestlé’s CIO. Joining the Initiative will help them learn alongside other leading organizations about evidence-based patterns for human-led, responsible AI. 

Frontier focus: Use AI at scale to transform the value chain and work smarter, better, and faster. 

AI bona fides: Nestlé has already embedded AI into core operations, creating new products, analyzing supplier contracts, and powering logistics. Digital twins in factories improve efficiency and sustainability, while everyday AI—including Copilot—is already boosting productivity for nearly 100,000 employees, more than one-third of Nestlé’s workforce globally.