Although the term 'Generative AI' (GenAI) is widely recognized, its practical application in daily workflows has yet to be understood. This exercise introduces students to GenAI tools, demonstrating how they can be seamlessly integrated into professional work practices to co-invent, analyze data, generate images, summarize text, etc. The exercise guides students through developing a fictional snack company, showcasing the versatility of GenAI in tasks such as market analysis, brand development, and the formulation of marketing strategies. The exercise culminates with students creating a comprehensive design document and presentation that can be used to pitch to investors. Key takeaways include understanding how GenAI works, incorporating AI into developing and launching a new product, and offering valuable lessons on AI's potential and limitations in the modern workplace.
In 2008 Abbott introduced a revolutionary new device, FreeStyle Navigator, designed to improve the lives of the world's more than 500 million diabetes patients by offering continuous glucose monitoring with technology that could translate an electrochemical signal from the body into precise, real-time data. Doctors and patients who tried the device appreciated it-but it was bulky, hard to manufacture, and expensive. And without widespread adoption, it wouldn't have the hoped-for impact. Abbott's leaders soon realized that they needed to go back to the drawing board. Four years later they launched FreeStyle Libre, a reimagined continuous glucose monitoring (CGM) system in which an even smaller sensor applied to a patient's arm sends data directly to a smartphone app every minute. It is now used by millions globally, and by the end of 2024 it will have generated more than $6 billion in revenue, making it one of the most successful medical devices-as measured by usage and sales-in history. The pivot from the Navigator to the Libre was a deeply consequential decision for Abbott, and others can learn from the principles the company's leaders followed to arrive at and then execute on that choice.
Generative artificial intelligence (gen AI) has the potential to radically alter how business is conducted, and there's no doubt that it will create a lot of value. Companies have used it to identify entirely new product opportunities and business models; to automate routine decisions, freeing humans to focus on decisions that involve ethical trade-offs, empathy, or imagination; to deliver customized professional services formerly available only to the wealthy; and to develop and communicate product and other recommendations to customers faster, more cheaply, and more informatively than was possible with human-driven processes. But, the authors ask, will companies be able to leverage gen AI to build a competitive advantage? The answer, they argue in this article, is no-unless you already have a competitive advantage that rivals cannot replicate using AI. Then the technology may serve to amplify the value you derive from that advantage.
Many experts are urging established companies to radically innovate-and disrupt themselves before someone else does. The trouble is, large firms aren't designed for moon shots. Their owners don't like the risks and won't kill the goose that lays the golden egg. As a result, all too often they end up defaulting to incremental innovation. But there is a solution: Incumbents can partner with entrepreneurial start-ups or with intrapreneurs that have ideas for breakthrough products. By doing that, they can leverage their significant resources while increasing the odds that those ideas will take off. This approach does require careful management, however. Drawing on the experiences of more than a dozen large multinationals, including Atlas Copco, Enel, and Epiroc, this article outlines a three-stage innovation process for incumbents to follow: First, set up numerous projects with multiple partners, nurturing them until their chances of success become clear. Next, once a venture has a breakthrough, gradually increase your commitment and help it remove roadblocks. Finally, when its business model is viable and it has a critical mass of customers, rapidly mobilize the resources it needs to scale up quickly.
Marketers have begun experimenting with AI to improve their brand-management efforts. But unlike other marketing tasks, brand management involves more than just repeatedly executing one specialized function. Long considered the exclusive domain of creative talent, it encompasses multiple activities designed to build the reputation and image of a business-such as crafting and communicating the brand story, ensuring that the product or service and its price reflect the brand's competitive positioning, and managing customer relationships to forge loyalty to the brand. A brand is a promise to customers about the quality, style, reliability, and aspiration of a purchase. AI can't fulfill that promise on its own (at least not anytime soon). But it can shape customers' impressions of a brand at every interaction. And it can automate expensive creative tasks-including product design. To succeed with it, you must understand how it is perceived by stakeholders and what can be done not only to mitigate their concerns but to make them avid supporters. Using examples from Intuit, Caterpillar, and LOOP, along with in-depth scholarly research, the authors propose a framework for thinking about the key roles that AI plays when it comes to managing brands effectively.
Today artificial intelligence can be harnessed by nearly anyone, using commands in everyday language instead of code. Soon it will transform more than 40% of all work activity, according to the authors' research. In this new era of collaboration between humans and machines, the ability to leverage AI effectively will be critical to your professional success. This article describes the three kinds of "fusion skills" you need to get the best results from gen AI. Intelligent interrogation involves instructing large language models to perform in ways that generate better outcomes-by, say, breaking processes down into steps or visualizing multiple potential paths to a solution. Judgment integration is about incorporating expert and ethical human discernment to make AI's output more trustworthy, reliable, and accurate. It entails augmenting a model's training sources with authoritative knowledge bases when necessary, keeping biases out of prompts, ensuring the privacy of any data used by the models, and scrutinizing suspect output. With reciprocal apprenticing, you tailor gen AI to your company's specific business context by including rich organizational data and know-how into the commands you give it. As you become better at doing that, you yourself learn how to train the AI to tackle more-sophisticated challenges. The AI revolution is already here. Learning these three skills will prepare you to thrive in it.
Marketplaces are the quintessential type of business that can profit from network effects: The greater the number of buyers who join one, the more attractive it becomes to sellers, and vice versa. Indeed, marketplaces such as Amazon, Booking.com, and Apple's App Store have achieved some of the strongest competitive positions imaginable. That's why entrepreneurs are seeking to build, and venture capitalists are seeking to invest in, the next Airbnb, Uber, or Twitch. But not all marketplaces have the potential to realize strong network effects-the kind that make a marketplace defensible against wannabe competitors. Differentiation among sellers, fragmentation of the seller and buyer bases, the value added by discovery and transaction services, and the importance of seller ratings are all decisive in determining whether a marketplace can flourish. That's why it is extremely important, both for new and established companies trying to build marketplaces and for their potential investors, to dig deeper into those factors. Drawing on their more than two decades' worth of research and their own experience as angel investors in some 40 marketplace start-ups, the authors offer a comprehensive list of questions that anyone thinking of launching a marketplace should explore.
Too many companies are failing to innovate. One reason, say the authors, is the polarized approach companies take to innovation. At one end of the spectrum, corporate R&D efforts tend to focus on product refreshes and incremental line upgrades that generate modest growth for lower risk. At the other end, venture capitalists favor high-risk "transformational" innovations that seek to upend industries and generate outsize returns. But there's a better, middle, way. This article presents the growth driver model, a framework that partners corporations with outside investors to identify and develop innovation opportunities, drawing on corporate resources and talent and externally recruited entrepreneurs. The authors illustrate the model with a detailed case study of how it revived innovation at Cordis, a large medical technology device maker.
Four new books explore how AI can help-and hinder-our productivity: Co-Intelligence, by Ethan Mollick; The Singularity Is Nearer, by Ray Kurzweil; The Mind's Mirror, by Daniela Rus and Gregory Mone; and Slow Productivity, by Cal Newport.
Pursuing decarbonization goals can feel like a frustratingly slow marathon with hidden curves. But leaders can learn by observing printer's running on the same track they are. The fast-paced decarbonization effort by the sustainability team of the 2024 Paris Olympics and Paralympics is illuminating. As they raced to reduce greenhouse gas emissions by half compared with recent Games, the team had to learn on the fly. Here are key takeaways for organizations on their own sustainability journeys.
Alexander and Lorenzo were at a crossroads with their biotech startup, PhagoMed. They had left high-profile positions at Boston Consulting Group (BCG) three years earlier, driven by a bold vision to tackle the global crisis of antibiotic resistance using phages, viruses with the remarkable ability to target specific bacteria. Despite a robust commitment to R&D yielding deeper insights into phage biology, the journey from lab to clinic was frustratingly slow. And with their capital dwindling, they faced the urgent need to reassess their strategic direction and resource allocation.
Many companies have large stores of customer data that can be tapped for valuable insights via analytics. At the same time, cybersecurity tactics used to protect personal information within that data can render it less useful for analysis. Data science practices will increasingly require that teams collaborate with IT on each use case to identify which techniques will maximize data privacy while still exposing useful information in the data set for analysis.
Advanced analytics, such as predictive and prescriptive models to support business decisions, remain the primary drivers of data science value in the enterprise. How might the flashy, fluent, but not entirely reliable generative AI large language models contribute to traditional analytics practice? The author describes some experimental prompts that show potential for labeling data and explaining model predictions, and shares guidance on monitoring and verifying that output.
It is well known that a significant reason machine learning projects fail to deliver business value is data scientist's failure to adequately understand the business context. Development teams can avoid mistakes when they put aside any reticence to ask basic questions and engage with colleagues on the business side. The authors advise gaining input from all involved stakeholders and suggest some specific types of queries that might help ML developers get to the heart of the problem at hand.
Algorithmic auditing aims to identify and monitor potential harms caused by algorithmic systems. In these scenarios, diverse stakeholders consider specific use cases and collaborate to address critical questions about who could be harmed by such technologies, and how. The authors present two tools " the Ethical Matrix and the Explainable Fairness framework " that can help organizations identify these potential harms.
Having a human in the loop is critical to mitigating the risks of generative AI errors and biases. But humans are also vulnerable to errors and biases and may trust artificial intelligence either too much or not enough. Findings from a field experiment by MIT and Accenture suggest that targeted friction in the form of labels that flag potential errors and omissions can direct users' attention to content that should be given closer inspection without sacrificing efficiency.
Keeping humans in the loop with AI systems is meant to mitigate concerns about unintended consequences of AI systems and facilitate intervention when those systems make questionable recommendations. However, ongoing research is finding that when using an automated system, humans often fail to engage their sense of responsibility in favor of trusting the AI.
On a day in early June 2017, Fadi Kanaan, a member of the organizing committee of the Beirut International Model United Nations (BEYMUN) conference at the American University of Beirut (AUB), sat in front of college hall feeling very excited yet worried. He had just received an email informing him that he would be the secretary-general of the annual Model United Nations conference at AUB in the spring of 2018. BEYMUN invited students from universities worldwide who were interested in addressing global concerns and prevailing world topics in a real-life simulation of the United Nations committee sessions. This conference was the perfect opportunity for participants to tackle such topics as regional conflicts, women and children, human rights, peacemaking, disarmament, economic and social development, and the environment while working in committees for Disarmament and International Security, Economic and Financial, Social Humanitarian and Cultural, and Special Political and Decolonization. But Kanaan was worried how BEYMUN would manage the overwhelming application process. Previously, the BEYMUN team had experienced issues with organizing large events, and Kanaan realized the existing system that received applications was inadequate for processing the hundreds of applications he expected the conference to receive. He knew that within two weeks of announcing the conference dates, applications would start to pour in. Kanaan and the team had to respond to each application before a deadline. Now was the time for BEYMUN to develop a system that would efficiently manage and process the applications to help organize a successful conference.
In fall 2023, the Food Program of Met Council-America's largest Jewish charity dedicated to fighting poverty-completed the rollout of the newest version of its digital pantry platform to twelve food pantries in the Met Council food pantry network. The digital initiative coincided with a shift from food pantries' traditional "pre-packed" model-in which pantry staff and volunteers pre-packed standardized bags of foods and handed them out to long lines of waiting clients (the standard model in the US)-to a "client choice" model, where clients could choose their own food items. Over half of the pantries in Met Council's network were undergoing the transition to client choice. For most of these pantries, the client choice model was initially implemented as an in-person shopping experience, similar to a small-scale grocery store. For the digital pantries, though, clients would be able to see available items and place orders online, similar to an online grocery shopping experience. Met Council viewed the digital initiative as the next step towards increasing the dignity of the pantry experience and incentivizing healthy food choices. This case discusses the evolution of the digital pantry; specifically, the pros and cons of each pantry model from an operational efficiency perspective, how operational levers can influence consumers' purchasing decisions, fairness in resource allocation problems, and "push" versus "pull" inventory distribution models.