As organizations increasingly rely on agentic AI, three structural shifts will require leaders to proactively address decision rights allocations, power, and decision-making practices.
By Michael Schrage and David Kiron
Incorporating AI agents at scale transforms how enterprises define, design, and deploy decision environments.1 Our research shows that organizations using AI to generate sophisticated choice sets — rather than singular, “best,” or “optimal” solutions — achieve superior outcomes across diverse sectors. These intelligent systems don’t just enhance decision-making; they push organizations to redesign decision rights, accountability frameworks, and power dynamics among decision makers.
Building on choice architecture principles from behavioral economics, our intelligent choice architecture (ICA) framework captures how such increasingly sophisticated systems reshape enterprise decision-making.2 (See “Intelligent Choice Architectures.”) By combining generative and predictive AI capabilities to create, refine, prioritize, and present options, ICAs transcend conventional recommendation engines. As AI agents, ICAs can articulate and explain trade-offs, surface hidden opportunities, and learn from outcomes to refine future choice sets. ICAs mark a decisive shift from using algorithms primarily for task automation efficiencies to deploying AI as an architect of superior decision environments.
Intelligent Choice Architectures (ICAs)
Intelligent choice architectures are dynamic systems that combine generative and predictive AI capabilities to create, refine, prioritize, and present choices for and with decision makers. They actively learn from outcomes, seek information, and alter the options available to decision makers.
Consider a major retail enterprise whose HR department deploys AI to identify emerging talent in sales and merchandising — a strategic imperative. While the AI system proves adept at spotting high-potential candidates in unexpected corners of the organization, it quickly becomes clear that these exceptional workers will remain underappreciated without a new decision rights frame- work to govern development, transfer, and promotion decisions. Success requires the implementation of a data- driven, collaborative decision architecture that aligns talent development options with management incentives, organizational priorities, and concrete outcomes. The lesson is clear: Unlocking the value potential of internal talent requires reallocating decision rights.
As AI capabilities evolve, ICAs will progress beyond mere decision support tools to become sophisticated augmentation systems for human decision makers. They will create powerful new frameworks wherein human judgment and machine intelligence work together to increase enterprise value creation.
The late Harvard Business School professor Michael C. Jensen devoted decades of research to determining how the distribution of decision rights drives corporate performance and what companies can do to allocate them most effectively. Jensen argued that decision rights allocation is “an extraordinarily difficult and controversial management task,” warning of the potential dangers of the over centralization and overdemocratization of decision-making.3
As compound AI systems — systems that combine predictive and generative AI — learn to become more sophisticated choice architects, enterprises’ focus shifts from decision execution to decision design. Executives become accountable for the decision environments in which staff members operate, including defining when AI-generated nudges must be acted upon and when they can be overridden. Just how empowering or constraining should the intelligent choice sets generated for executive and managerial decisions be? Consider, for example, a trading algorithm that discovers a novel market pattern. Should it wait for human validation before acting? What about an ICA agent managing supply chain operations that identifies a more efficient logistics strategy: What permissions are required before implementing it? Under what conditions should the organization encourage human initiative versus obedience and compliance? These are questions leaders must consider.
ICA agents should reflect and respect an organization’s values and aspirations. In the Decision Rights 2.0 era, enterprises must determine who has the authority and responsibility to architect, deploy, and govern choice environments where human judgment and AI capabilities intersect. This authority carries explicit accountability both for immediate outcomes and the long-term effectiveness of decision architectures. This AI-driven redefinition elevates decision rights from a set of enterprise rules and practices regarding who can make specific decisions, what they can decide, and how strategic decisions shape how organizations harness the combined power of human judgment and artificial intelligence.
Indeed, ICA agents don’t just provide decision support — they create decision environments in which superior choices emerge from the interplay of machine intelligence and human judgment. Think of commercial aviation flight management systems that advise pilots: They don’t simply process navigation data; they preserve flight-path data and adapt to different routes, weather patterns, and pilot preferences, all while operating within strict safety parameters. Similarly, enterprise ICA agents continuously learn while operating within clear operational, legal, and regulatory boundaries. This directly addresses the too-common fear that ever-smarter and more capable AI systems will render human judgment marginal and/or irrelevant. In fact, the opposite is true. As ICA agents take on the heavy lifting of data analysis, pattern recognition, and optimization, they free their human counterparts and collaborators to focus on higher-order challenges.
Liberty Mutual effectively created an ICA agent to help train new claims adjusters by giving them more tailored training based on 20,000 company knowledge articles. This ICA agent helps adjusters more efficiently triage incoming customer calls to quickly resolve inquiries. The AI agent is one implementation of GenAI across the company. Additionally, one year after the companywide deployment of LibertyGPT, the organization’s internal instance of OpenAI’s ChatGPT, Liberty Mutual has seen it improve and support internal employee productivity. Liberty Mutual has saved more than 200,000 person-hours compared to previous settlements work- loads, says Monica Caldas, the company’s global chief information officer.
With ICAs, significant corporate decisions depend as much on the nature and purpose of intelligent decision environments as they do markets, products, culture, or strategy. A new focus on meta-decision rights emerges: the design and governance of the systems generating choices. A new meta-decision imperative requires human leadership teams and intelligent algorithms to come together to determine how decision rights around decisions rights should be effectively allocated by the human leaders and the cutting-edge algorithms responsible for those rights. Ironically, leaders seeking to maximize value from AI have little choice but to meet these meta-decision obligations.
We see three structural shifts taking place that will require leaders to proactively address decision rights allocations, power, and decision-making practices in the agentic AI enterprise.
Power Flows to Intelligent Choice Architects — Humans and Machines
As ICAs gain traction in the enterprise, the focus of decision rights will shift from those who make decisions to those who design better decision environments. This is the case at French multinational pharmaceutical company Sanofi, where ICAs are systematically helping to enhance the decision-making process. The company’s research and innovation teams have adopted data-driven strategies to optimize project investments and overcome challenges like sunk-cost bias, a common industry pitfall that can make it difficult to pivot from underperforming assets. Sanofi’s AI systems now equip leaders to confidently redirect resources based on data-driven insights. This transformation in decision-making showcases the power of well-designed ICAs to amplify human judgment and ensure that decisions are aligned with enterprise goals. As Sanofi chief digital officer Emmanuel Frenehard observes, “We deploy AI first and foremost to systematically reduce human cognitive biases.” Ensuring that portfolio decisions are guided by data rather than emotional attachments or past expenditures is a key function of Sanofi’s ICAs.
Network Effects Amplify and Enhance Decision Intelligence
ICAs create network effects whereby each decision improves the system’s predictive accuracy and future decision quality. The ICAs at multinational payments technology company Mastercard, such as those that proactively identify transaction patterns to prevent fraud and address unnecessary card declines, exemplify this virtuous cycle. Every transaction that flows through the system enriches the ICAs’ effectiveness, leading to better anomaly detection and enhanced customer experiences. Leaders design decision environments to continuously refine themselves as they work, fueling network effects. This ensures that the ICAs’ learning capabilities benefit the entire organization (not to mention merchants, card issuers, and cardholders). “With more data, we add more value to our services,” explains chief data and artificial intelligence officer Greg Ulrich. “With more services, we add value to our payments ecosystem.”
We see three structural shifts taking place that will require leaders to proactively address decision rights allocations, power, and decision-making practices.
Real-Time Optimizations Redefine Authority and Oversight
Algorithmic trading and programmatic advertising illustrate how ICA agents can autonomously and dynamically refine tactics to maximize outcomes at a pace that exceeds what human managers can achieve. Ceding decision-making authority to these agents in these contexts increases the likelihood of beneficial outcomes. We also see this phenomenon with revenue forecasting. At one pharmaceutical company, an ICA agent was used to generate revenue forecast alternatives. As the model evolved, the most accurate forecasts were generated from collaborations between finance managers and the ICA agent. That is, the ICA agent evolved from a decision-support tool to a collaborator. Eventually, the ICA agent became even more accurate without human intervention. In this context, the quasi-autonomous agent gained the authority to set revenue projections and dynamically adjust them as new information emerged.
These three shifts call for a fundamental rethinking of enterprise power structures. Success in the AI era belongs to human executives who are willing to collaboratively build and orchestrate intelligent choice environments that ensure that human judgment remains appropriately engaged in value creation. The question for leaders charged with organizational design will shift from “Who should decide?” to “How do we architect better ways to decide?”
DECISION RIGHTS FOR THE AGENTIC AI ERA
Organizations must rebuild decision rights frameworks around three core principles:
Anticipating Power Conflicts
Pragmatically, ICAs aren’t limited to being tools that inform human decisions; they are entities that can learn to develop decision-making capabilities themselves. These capabilities present a fundamental governance challenge: As ICAs learn and improve, they evolve from mere decision-support tools into increasingly capable decision makers that may match or even exceed expert human performance.
When an ICA’s ability to make effective decisions exceeds its formally granted decision rights, this creates an inherent conflict: the learning-authority dilemma. To address this predicament and still harness the full potential of these intelligent systems, organizations need dynamic governance frameworks that systematically evaluate ICA capabilities and intentionally expand their authority when warranted, while maintaining appropriate oversight and controls. Striking a balance between benefiting from enhanced capabilities and maintaining responsible governance as ICAs evolve beyond their initial decision constraints will become a leadership imperative for deploying AI agents at scale. (See “Decision Rights for the Agentic AI Era.”)
As AI becomes more sophisticated at architecting choices for and with humans, the most critical decision rights will center on meta-decisions — choosing how to design the systems that make choices. This will create a new hierarchy of decision rights, where the power to shape decision environments supersedes the power to make individual decisions. Leaders are becoming accountable not just for decisions made but for the quality of the ICAs they create.
Some actions that leaders can take as they move in this new direction include the following:
GETTING AHEAD OF GOVERNANCE FOR AI-DRIVEN CHOICE SYSTEMS. Companies should establish ethical, strategic, and operational guardrails for their ICAs.
ELEVATING COGNITIVE CONTRIBUTIONS. This requires a shift from tactically enabling decision-making to overseeing the design of intelligent decision environments.
ENSURING ALIGNMENT. Leaders must create systems and processes to ensure that AI-generated decisions further organizational values and goals.
ESTABLISHING NEW METRICS. These should measure choice quality and diversity, decision environment effectiveness, learning and adaptation rates, and network intelligence optimization.
The future belongs not to those who make the best decisions but to those who create the best decision environments.
1. In the context of this research, decision environments are dynamic settings in which people and algorithms consider complex information, competing priorities, and constraints to make consequential decisions.
2. M. Schrage and D. Kiron, “Intelligent Choices Reshape Decision-Making and Productivity,” MIT Sloan Management Review, Oct. 29, 2024, https://sloanreview.mit.edu.
3. Jensen, M. “Decision Rights: Who Gives the Green Light?” Harvard Business School Working Knowledge, Aug. 8, 2005, www.library.hbs.edu.
THE RESEARCH
This second article in the 2024-2025 strategic measurement series, conducted in collaboration with Tata Consultancy Services, examines how leading organizations integrate predictive and generative AI to develop and present improved choices to human decision makers. Drawing on interviews with senior leaders in six major industry groups, our research reveals the emergence of intelligent choice architectures — a new paradigm where AI systems proactively participate in structuring and shaping strategic decisions. The implications for organizational performance, decision rights, and strategic agility are significant, particularly as businesses navigate increasing complexity and compressed decision cycles.
AUTHORS
Michael Schrage is a research fellow with the MIT Sloan School of Management’s Initiative on the Digital Economy.
His research, writing, and advisory work focuses on the behavioral economics of digital media, models, and metrics as strategic resources for managing innovation opportunity and risk.
David Kiron is the editorial director, research, of MIT Sloan Management Review and program lead for its Big Ideas research initiatives.
CONTRIBUTORS
Sankaranarayanan Viswanathan
Vice president, Corporate Technology Office, Tata Consultancy Services
ACKNOWLEDGMENTS
We thank each of the following individuals, who were interviewed for this article:
Monica Caldas
Global CIO, Liberty Mutual Insurance
Emmanuel Frenehard
Chief Digital Officer, Sanofi
Greg Ulrich
Chief Data and Artificial intelligence officer, Mastercard
In banking, financial services, and insurance — industries where decisions can influence the behavior of millions of customers — developing better choices represents a strategic, operational, and organizational imperative.
In the media industry and capital-intensive industries like communications and technology, business leaders are under relentless pressure to optimize operational efficiency, improve decision-making, and unlock new revenue opportunities.
As AI-driven decision environments evolve, the question for energy, resources, and utilities leaders is not whether to automate but how to architect decision-making and decision rights.
Life sciences and health care quality and outcomes fundamentally depend on the quality of decisions made across increasingly interconnected ecosystems of providers, payers, pharmaceutical companies, drug distributors, pharmacy benefit managers, medical technology makers, and others that these broad sectors comprise.
As increasingly capable AI systems generate thousands of production design variations, predict maintenance needs, and optimize supply chains, the fundamental opportunity for manufacturing leaders isn’t “automation” or better harnessing intelligent systems to improve business process efficiencies.
Intelligent choice architectures are a natural next step to providing the adaptive intelligence necessary to navigate increasingly complex retail and CPG business environments.
