AI Agents Aren’t Ready for Consumer-Facing Work—But They Can Excel at Internal ProcessesSKIP TO CONTENTHarvard Business Review LogoHarvard Business Review LogoGenerative AI|AI Agents Aren’t Ready for Consumer-Facing Work—But They Can Excel at Internal ProcessesSubscribeSign InLatestMagazineTopicsPodcastsStoreReading ListsData & VisualsCase SelectionsHBR ExecutiveSearch hbr.orgCLEARSubscribeLatestPodcastsThe MagazineStoreWebinarsNewslettersAll TopicsReading ListsData & VisualsCase SelectionsHBR ExecutiveMy LibraryAccount SettingsSign InExplore HBRLatestThe MagazinePodcastsStoreWebinarsNewslettersPopular TopicsManaging YourselfLeadershipStrategyManaging TeamsGenderInnovationWork-life BalanceAll TopicsFor SubscribersReading ListsData & VisualsCase SelectionsHBR ExecutiveSubscribeMy AccountMy LibraryTopic FeedsOrdersAccount SettingsEmail PreferencesSign InHarvard Business Review LogoGenerative AIAI Agents Aren’t Ready for Consumer-Facing Work—But They Can Excel at Internal Processes by Nathan Furr, Jur Gaarlandt, Sid Mohan and Andrew ShipilovNovember 25, 2025piranka/Getty ImagesPostPostShareSavePrintSummary.   Leer en españolLer em portuguêsPostPostShareSavePrintOver the past two years, we’ve all heard a lot about generative AI. But for all of the hype about how it might change the business world, finding concrete examples of what it is doing right now isn’t always easy. More often, companies report struggling—and failing—to create value with their AI experiments.Nathan Furr is a Professor of Strategy at INSEAD and a coauthor of five best-selling books, including The Upside of Uncertainty, The Innovator’s Method, Leading Transformation, Innovation Capital, and Nail It then Scale It.JGJur Gaarlandt is the Partner Benelux at Artefact.SMSid Mohan is the Director of Data Science and AI for Artefact Northern Europe and the U.S. Sid brings significant experience from having deployed numerous AI and agentic AI solutions for numerous U.S. and European clients.Andrew Shipilov is a John H. Loudon Chaired Professor of International Management at INSEAD. He is a coauthor of Network Advantage: How to Unlock Value From Your Alliances and Partnerships.PostPostShareSavePrintRead more on Generative AI or related topics Technology and analytics, AI and machine learning, Automation and Enterprise computingPartner CenterStart my subscription!Explore HBRThe LatestAll TopicsMagazine ArchiveReading ListsCase SelectionsHBR ExecutivePodcastsWebinarsData & VisualsMy LibraryNewslettersHBR PressHBR StoreArticle ReprintsBooksCasesCollectionsMagazine IssuesHBR Guide SeriesHBR 20-Minute ManagersHBR Emotional Intelligence SeriesHBR Must ReadsToolsAbout HBRContact UsAdvertise with UsInformation for Booksellers/RetailersMastheadGlobal EditionsMedia InquiriesGuidelines for AuthorsHBR Analytic ServicesCopyright PermissionsAccessibilityDigital AccessibilityManage My AccountMy LibraryTopic FeedsOrdersAccount SettingsEmail PreferencesAccount FAQHelp CenterContact Customer ServiceExplore HBRThe LatestAll TopicsMagazine ArchiveReading ListsCase SelectionsHBR ExecutivePodcastsWebinarsData & VisualsMy LibraryNewslettersHBR PressHBR StoreArticle ReprintsBooksCasesCollectionsMagazine IssuesHBR Guide SeriesHBR 20-Minute ManagersHBR Emotional Intelligence SeriesHBR Must ReadsToolsAbout HBRContact UsAdvertise with UsInformation for Booksellers/RetailersMastheadGlobal EditionsMedia InquiriesGuidelines for AuthorsHBR Analytic ServicesCopyright PermissionsAccessibilityDigital AccessibilityManage My AccountMy LibraryTopic FeedsOrdersAccount SettingsEmail PreferencesAccount FAQHelp CenterContact Customer ServiceFollow HBRFacebookX Corp.LinkedInInstagramYour NewsreaderHarvard Business Review LogoAbout UsCareersPrivacy PolicyCookie PolicyCopyright InformationTrademark PolicyTerms of UseHarvard Business Publishing:Higher EducationCorporate LearningHarvard Business ReviewHarvard Business SchoolCopyright ©2025 Harvard Business School Publishing. All rights reserved. Harvard Business Publishing is an affiliate of Harvard Business School.

Generative AI’s rapid emergence has triggered widespread speculation regarding its transformative potential across numerous industries, yet the practical application of this technology within business operations has, to date, largely failed to deliver on initial expectations. Numerous companies have reported difficulties in realizing value from their AI experiments, suggesting a disconnect between the theoretical promise of the technology and its tangible outcomes. This article, authored by Nathan Furr, Jur Gaarlandt, Sid Mohan, and Andrew Shipilov, critically examines this situation, arguing that while consumer-facing applications of generative AI are currently ill-suited, the technology possesses significant potential for optimizing internal business processes.

The core argument centers on a recognition of the current limitations of generative AI when deployed directly to engage with external stakeholders—customers, clients, or the public. The authors posit that the inherent complexity of understanding nuanced human intent, rapidly evolving context, and the need for adaptive responses effectively preclude the robust performance of generative AI in these consumer-facing scenarios. The technology’s ability to generate text and images, while impressive, isn't sufficient to replicate the critical elements of human interaction, particularly regarding empathy, judgment, and the ability to navigate ambiguous situations. This is supported by the observation of widespread experimentation failures, indicating a fundamental misalignment between the ambitious goals and the current capabilities of generative AI.

Conversely, the article makes a compelling case for the application of generative AI within internal organizational functions. The authors suggest that the technology’s strengths—its capacity for rapid analysis, pattern recognition, and automated content creation—are ideally suited for streamlining and enhancing processes that operate primarily within a company's own ecosystem. Specifically, they identify several areas where generative AI can provide demonstrable value. These include accelerating data analysis to identify operational bottlenecks, automating the creation of reports and documentation, improving knowledge management systems by extracting key insights from vast datasets, and assisting in the development of strategic initiatives through scenario planning and predictive modeling.

The article emphasizes a strategic shift in thinking, advocating for focusing on internal efficiency and operational excellence rather than attempting to replicate human interaction in external-facing roles. This approach aligns with the authors’ broader work on innovation capital and transformation, suggesting that companies should invest in leveraging technology to augment human capabilities rather than seeking to replace them. The deployment of generative AI within internal processes is presented as a means to unlock greater productivity, reduce operational costs, and improve decision-making.

Furthermore, the piece subtly suggests that the hype surrounding generative AI’s broad applications has driven unrealistic expectations, causing premature investment and, consequently, unsuccessful outcomes. By refocusing attention on the technology’s more constrained, yet achievable, applications within internal processes, the authors advocate for a more grounded and pragmatic approach to AI adoption. The ultimate goal, as articulated by Shipilov and Furr through their prior research, is not merely technological advancement, but rather the creation of sustainable competitive advantage through intelligent operational design. This necessitates a methodical approach, starting with well-defined problems within the organization and using generative AI to systematically improve them, an iterative process aligned with the concepts outlined in their “Innovator’s Method” and “Nail It Then Scale It” frameworks. The authors’ combined expertise, drawing from their research on network advantage and transformation, strengthens the argument that AI should be strategically integrated within the broader context of organizational capabilities and relationships, offering a nuanced perspective on the future of AI within business.