CISR Glossary

A bounded autonomous system that uses observations from its environment to select and execute actions in pursuit of specified goals. In contemporary business applications, this is often implemented using a large language model that orchestrates tool use and action sequences under defined constraints. AI agents can adapt based on experience

A set of skills required to effectively instruct GenAI tools and evaluate its output. AIDE skills include the ability to write prompts with precise problem statements, context-specific objectives, and detailed process descriptions; verify the accuracy and relevance of the generated content; and evaluate whether one’s use of the tool is consistent with organizational policy

The ability to manage AI initiatives in ways that ensure models are value-generating, compliant, representative, and reliable

The organizational ability to sustain large numbers of interdependent, production AI models over time

Embedding and monitoring compliant, human-centered, and transparent AI practices by design

Aligning AI investments with strategic goals and offering measurable, scalable value

Creating AI-ready people, roles, and teams while redesigning work around AI capabilities

Architecting modular, interoperable platforms and data ecosystems to enable enterprise-wide intelligence

Applied analytics models that have some level of autonomy. AI includes techniques such as machine learning, natural language processing, and generative AI

The gradual expansion of what AI agents are authorized to do as organizations delegate tasks incrementally, without revisiting who oversees these actions or who is accountable for their consequences

The practice of employees using publicly available GenAI tools for work purposes without organizational oversight or approval. BYOAI risks can include data loss, intellectual property leakage, copyright violation, and security breaches

Three different strategies that organizations can use to develop GenAI solutions. “Buy” refers to purchasing ready-made solutions; “boost” involves enhancing vendor AI models with proprietary data to create the solution; and “build” involves either training a custom GenAI solution from the ground up, or fine-tuning an open-source foundation model to create the solution

An AI-enabled system that collaborates with humans to perform complex work 

Risks inherent to the generative AI technology and models that an organization adopts. These risks exist before the organization makes any deployment decisions and are shaped by training data quality, model behavior, and performance drift introduced by vendor updates

Risks that emerge from choices organizations make about how they deploy, configure, and use generative AI. These choices include system prompt design, data handling, security safeguards, verification requirements, agent permissions, and use contexts

A type of AI technology trained on vast amounts data that can be adapted to perform a wide variety of tasks. These models can be fine-tuned for specific domains or tasks by incorporating smaller, specialized, curated datasets into the training set or by using software plug-ins or add-ons. Large language models (LLMs) are general-purpose foundation models

A type of artificial intelligence that uses neural network architectures to identify patterns in diverse, extensive, voluminous training data, allowing it to quickly and autonomously generate new content in the form of text, images, audio, and video

The set of components involved in a GenAI implementation that give rise to risk, including training data, foundation models, prompts, outputs, and use decisions. Retrieval-augmented generation (RAG) and AI agents compound exposure through data connections, autonomous actions, and agent-to-agent interactions

The set of components involved in a GenAI implementation that give rise to risk, including training data, foundation models, prompts, outputs, and use decisions. Retrieval-augmented generation (RAG) and AI agents compound exposure through data connections, autonomous actions, and agent-to-agent interactions

Targeted generative AI applications that are developed to meet specific organizational needs and strategic business objectives. GenAI solutions are designed for use by specific groups of organizational stakeholders and integrate into processes, systems, and offerings to create value at scale

Broadly applicable generative AI applications designed to enhance personal productivity across various roles and functions. They include conversational AI systems and digital assistants embedded in existing productivity software. Versatile and multipurpose, their use is generally not predefined, but instead identified and refined by users

A type of generative AI model capable of language generation and other natural language processing tasks. LLMs are trained on vast amounts of text data and can produce coherent and contextually relevant responses to user prompts by predicting the most likely sequence of words in a given context

AI models that are difficult to trace from input to output

The AI infrastructure developed in house. For instance, an AI platform developed in house that integrates purchased large language models and enterprise assets such as APIs, data, and business processes is proprietary AI

A technique used to enhance the performance of GenAI solutions by combining generation with data retrieval. Instead of relying solely on pre-existing data stored within a model, RAG retrieves relevant data from external sources, such as company databases or proprietary documents, and incorporates it into the model’s response. This approach improves the accuracy, specificity, and relevance of the generated output

Growing the value created by both a core trained model and recontextualized adaptations of the model

The use of GenAI tools or solutions within an organization by individual stakeholders or teams without formal oversight or authorization. Shadow GenAI can introduce risks, such as inconsistent outputs or data security vulnerabilities, due to a lack of organizational control

A type of AI technology, typically trained on smaller data sets, that is tailored to specific industries, tasks, or domains, and contains far fewer parameters than foundation models

Rules-based AI that relies on statistical algorithms and machine learning to analyze data and make predictions. Also known as predictive AI

The use of people data and analytics in ways not just compliant with existing laws and regulations but also informed by organizational values and those of ecosystem actors

An organization’s ability to gather, store, and use data assets in ways that are compliant with existing laws and regulations and consistent with organizational and stakeholder values. Organizations with advanced acceptable data use capabilities have contextualized norms and policies. They have scalable oversight processes that ensure employees, partners, and customers appropriately engage with organizational data assets

The organizational ability to consistently communicate the CEO’s vision for data that motivates investment in data resources and data products. CEO-level data leadership, data value realization, and data lifecycle measurement are factors that amplify the financial impact of data monetization

Organizational arrangements that facilitate knowledge exchange between data experts and domain experts (see also “data-domain connections”)

Delighting customers and serving them cost efficiently, using a mix of wrapping and improving initiatives

The ability to gather accurate and actionable knowledge about customer needs and behaviors. Organizations with advanced customer understanding capabilities accurately grasp what customers need and value, can cocreate with customers, and can formulate and test hypotheses about customer preferences

Establishing a modular platform comprised of data objects

A cohesive set of data that is made readily usable

An organizational state whereby employees regularly and universally have (1) access to data assets, (2) the skills to exploit them, (3) the motivation to engage with them, and (4) guidance to use them strategically

The organizational ability to track and manage the creation of data assets and other data resources and their use across their lifecycle. CEO-level data leadership, data value realization, and data lifecycle measurement are factors that amplify the financial impact of data monetization

The ease of data asset recombination and reuse

Domain practice areas that function as levers to increase or limit data liquidity. There are three data liquidity levers: data architecture, data preparation, and data permissioning

The ability to produce data assets that people can find, use, and trust. Organizations with advanced data management capabilities can report on the accuracy of their data, match related data entries, consolidate and streamline data fields, and integrate related data from external sources such as data aggregators and suppliers

The generation of financial returns from data assets

Five capabilities required to execute data monetization initiatives

The extent to which an organization’s data initiatives help the organization realize its strategic business outcomes

A project to develop data solution products that offer data, insight, or action to employees, customers, technology, other data consumers, or a combination of these

A high-level plan that communicates how an organization will improve its bottom line using its data assets. A data monetization strategy is a component of an organization’s data strategy

The full set of resources that speed up data monetization initiatives, including data assets and data monetization capabilities. Data monetization capabilities may be found in people with expertise, or in expertise embedded in tools, routines, policies, forms, software, and so on

Establishing guardrails for acceptable data use

The ability to capture, transform, and disseminate data assets securely and efficiently. A data platform capability leverages contemporary, cloud-based software to ingest, process, secure, integrate, and deliver data assets. Organizations with advanced data platform capabilities can cost-effectively distribute data assets inside and outside of the organization at scale

Curating accurate data assets that are broadly relevant and strategically important for the enterprise

Data asset and data solution initiatives that aim to increase data asset liquidity and generate financial returns from data, respectively

A perspective organizations adopt for managing the market development, customer appeal, pricing, deployment, (re)use, continuous improvement, profitability, and retirement of their data assets and data solutions

The ability to use scientific methods, processes, algorithms, and statistics to extract meaning and insights from data assets. Organizations with advanced data science capabilities support data-savvy people across the organization in making evidence-based decisions. They leverage advanced statistics and techniques such as machine learning to inform and automate processes and products

An application, dashboard, website, system, or set of APIs that delivers specific functionality to users by consuming one or more data assets

Also know as the data-insight-action process, this occurs when people or systems use data to develop insights that inform action, which generates value

The organizational ability to move benefits created from data products to the organization’s bottom line, resulting in realized financial value. CEO-level data leadership, data value realization, and data lifecycle measurement are factors that amplify the financial impact of data monetization

The set of non-analytical, preparatory tasks required to make data usable. It encompasses the entire “friction” of data work—finding, permissioning, provisioning, preparing, and integrating data—that must be completed before any valuable insights can be drawn

Structures that facilitate knowledge exchange between data experts and domain experts

The process by which a company learns to run on analytics-powered business processes, fueled by employees who habitually use data to inform their work and by automation that draws upon data science techniques

A data asset that is valuable, modular, and accessible by way of a programmatic interface

A data monetization capability that is accessible organization-wide, not just by experts or people in certain pockets

An offering that combines deep domain knowledge with technology to deliver both content and workflow automation in order to drive improved outcomes and productivity for users

Enabling and encouraging the use of improving, wrapping, and selling initiatives, as needed, to improve customer experience while simplifying operations and generating novel business models and revenue streams

Leveraging data assets to solve problems for other organizations or consumer markets, using selling initiatives to produce information offerings and wrapping initiatives to sustain the sales of those offerings

A data monetization approach that generates money when organizations use data to change the economics of work for the better and then remove or redirect the resulting slack

The relationship between a provider and a customer when the two parties jointly create real value by integrating resources and drawing on shared knowledge in order to solve customer problems

A cohesive set of data that an organization has made accurate, available, combinable, relevant, secure, and readily usable for future value creation

Streamlining work tasks and eliminating inefficiencies, primarily from improving initiatives

A data monetization approach that generates new revenues when organizations commercialize data in the form of an information solution

When value that has been created from data is turned into money

A data monetization approach that generates money when organizations use data-fueled features or experiences to enhance the value proposition of a product and then raise prices or sell more products

A multi-faceted statement that articulates why the organization exists by outlining future aspirations, value propositions, and core values

A specification of who has the authority and accountability for key decisions in an organization

Enabling constraints by which empowered teams can operate with greater meaning, competence, direction, and impact

The five domains—principles, policies, people, processes, and platforms—through which organizations govern technology. Achieving minimum viable governance requires mechanisms across all five domains functioning as a mutually reinforcing system

A characteristic of minimum viable governance in which governance mechanisms function as a unified system

The least amount of governance required to manage risk effectively while enabling the organization to sense and seize opportunities. Minimum viable governance applies the philosophy of governing at the minimum level required across five governance domains (principles, policies, people, processes, and platforms)

A practice in which an employee takes on an additional job title and extra responsibilities without changing their original employment relationship, pay, and benefits

A specification of who has the authority and accountability for how to best achieve strategic goals

A characteristic of minimum viable governance in which governance accounts for the cost of inaction alongside the cost of action

A specification of who has the authority and accountability for what the organization needs to achieve and why

A characteristic of minimum viable governance in which governance mechanisms can be introduced, adjusted, or retired as conditions change

A characteristic of minimum viable governance in which oversight is built into governance mechanisms themselves

How a company makes money in the AI era. In the AI era, business models increasingly prioritize outcome-driven actions powered by autonomous, agentic AI systems. Our business model framework in the AI era differentiates four business models based on (1) how the company acts on behalf of customers (whether the company assists customers or represents them) and (2) whether the company takes a structured approach to business execution with AI (reliant on predefined processes) or an adaptive approach (reliant on outcomes and constraints).

How a company makes money in the digital economy. Our digital business model framework differentiates four digital business models based on (1) how well the company knows its end customer and (2) whether the company is designed as a value chain or ecosystem business. In our data collection from 2013 to 2025, companies’ dominant digital business models have gradually shifted from value chain to ecosystem.

A digital business model in which the company offers a go-to destination in a customer domain (e.g., buying and owning a house, managing energy use), connecting customers with providers

An AI-enabled business model in which the company augments its existing business model with AI, assisting customers in achieving outcomes by delivering products and services through AI-enhanced established processes

An AI-enabled business model in which the company assists the customer in achieving outcomes by using AI to adaptively assemble reusable, combinable modules, including those from other companies, into tailored service bundles that meet the customer’s goals

A digital business model in which the company provides a plug-and-play product or service (e.g., payments provider)

A digital business model in which the company combines digital and physical channels to sell directly to end customers (e.g., retailer, bank)

An AI-enabled business model in which the company represents the customer to autonomously achieve outcomes on their behalf through adaptive, AI-mediated collaboration across an ecosystem of products and services

A digital business model in which the company sells via other companies (e.g., manufacturer, insurance via an agent)

Distribution of responsibilities for digital offerings and components that balances autonomy and alignment

An approach to building and monetizing an external developer platform in which a company allows external developers to integrate data and functionality provided by the platform via APIs into their own, otherwise freestanding solutions. Often, these solutions combine data and functionality from multiple companies’ build-in external developer platforms.

The build-in approach is distinct from a build-on approach.

An approach to building and monetizing an external developer platform in which a company allows external developers to use its platform APIs to build solutions (such as apps) on the platform; these solutions often complement and extend the functionality of the company’s core offering, which is based on the digital platform underlying the external developer platform. A build-on external developer platform is often accompanied by a marketplace that connects external developers with company customers, allowing the external developers to distribute their solutions to customers.

The build-on approach is distinct from a build-in approach.

Information-enriched solution wrapped in a seamless, personalized customer experience

Repository of business, data, and infrastructure components used to rapidly configure digital offerings

Repository of digital components open to external parties. Abbreviated as ExDP.

A coherent set of standardized, integrated systems, processes, and data supporting a company’s core operations

The organizational ability to sense and seize opportunities that come from changes in the organization’s environment

Customer-facing roles with the responsibility to continuously make sense of customer needs and initiate change within the organization

Organizational learning about what customers will pay for and how digital technologies can deliver on their demands

An understanding, developed through experience and education, of the impact that emerging technologies will have on businesses’ success over the next decade

The extent to which the workspace, systems, social networks, and business rules adapt to work and how it is done. An adaptive work environment (1) connects people and ideas, (2) integrates activities and systems, and (3) governs with clarity and transparency

Shared behavioral norms that determine how employees collaborate across silos and hierarchies, innovate concerning work and customer initiatives, and make choices about work  

The physical, cultural, and digital arrangements that simplify working life in complex, dynamic, and often unstructured working environments

The extent to which employees of an organization are enabled or constrained by its adaptive work environment and collective work habits to do their jobs today and reimagine their jobs of tomorrow (i.e., how easy it is for employees to do their work of today and reimagine their work of tomorrow)

The process of analyzing employee data to quantify skills proficiency. The skills inference process consists of (1) defining a taxonomy of skills required to realize your organization’s purpose and strategic objectives, (2) gathering employee data as evidence of these skills, and (3) conducting an assessment of this evidence to quantify employees’ skill proficiency.

A structured classification of a wide range of skills organized into groups of capabilities and levels of proficiency, providing a comprehensive framework for understanding, assessing, and developing the skills an organization needs in order to realize its purpose and strategic objectives.

The discrepancy between the collective skills proficiency that an organization requires to achieve its strategic objectives and the current skills proficiency of its workforce.

The coordination model focuses on integration. A coordination model company may present a single face to its customers or create a transparent supply chain without forcing specific process standards on the company's operating units

The diversification operating model is a decentralized organizational design. The company unites to pursue different markets with different products and services, and benefits from local autonomy in deciding how to address customer demand

The organizing logic for business processes and IT infrastructure, reflecting the integration and standardization requirements of the company's operating model

The replication operating model focuses on process standardization. Operating units perform tasks the same way using the same systems so that they can generate global efficiencies and brand recognition. However, operating units rarely interact

The unification operating model describes a centralized organizational design. The company pursues the need for reliability, predictability, and low cost by standardizing business processes and sharing data across business units to create an end-to-end view of operations and present a single face to the customer

Producing operational processes digitized to increase efficiency, such as automated and standardized internal business processes

Producing new or improved services or solutions, such as services that complement products—e.g., predictive maintenance—and enriched customer experiences, such as ensuring that the services are available across multiple channels—e.g., mobile, web, call center—aimed at increasing revenue per product/service and per customer

The process of taking an idea that uses digital technologies from inception to impact

Producing digital tools or information aimed at greater employee productivity and retention, such as improvements in how people collaborate

Producing new sources of revenue from new customer segments, such as a luxury automobile manufacturer experimenting with mobility services for people who do not own a vehicle

Enabling multiple digital innovation initiatives to realize bottom-line value from their innovation by leveraging shared resources

An enterprise IT operating model that blends strong enterprise IT leadership with high innovation velocity, using unified data platforms and enterprise-grade AI to enable scalable, reusable digital capabilities. Technology strategy is federated: enterprise IT provides governance, security, interoperability, and reusable components, while business units drive localized innovation within controlled frameworks.

An enterprise IT operating model that empowers business units to drive rapid, customer-focused innovation in fast-changing environments. Enterprise IT orchestrates interoperability and security, while business units own and execute product and channel platforms. The model thrives on high innovation velocity, shared learning, and reuse across units.

An enterprise IT operating model that emphasizes enterprise-wide standardization to ensure compliance, reliability, quality, and cost control in slower-changing industries. Enterprise IT owns most platforms and data architecture, focusing on automation, integration, and risk reduction, while business units manage only the systems unique to their needs. Reuse across units drives efficiency and consistency.

An enterprise IT operating model describes the ways the enterprise organizes to create value from IT—the accountabilities, processes, platforms, metrics, and behaviors. The model describes the roles of the IT unit and the business units and how they collaborate. An effective enterprise IT operating model in the AI era enables the enterprise’s most valuable components and capabilities to innovate and grow and scales the use and reuse of data and AI. It also manages risks such as cyber threats, privacy breaches, supplier demands, and competitive threats from new business models.

An IT operating model defines how the IT organization creates, supports, and delivers value to the entire organization. The operating model includes the organizational structure and processes, decision rights, sourcing, funding models, and needed capabilities to create the necessary platforms, APIs, and enterprise architecture

An enterprise IT operating model used by mature organizations in slower-changing industries where business units retain most technology decision rights. Enterprise IT offers light governance and shared tools, while units optimize locally for their diverse needs. The model centers on modernizing legacy systems and selectively building reusable platforms but struggles to balance modernization with discipline.

Creating consistent, ongoing progress in digitally enabled business transformations by countering the sources of organizational inertia

A customer’s end-to-end need in an area such as home, mobility, energy, education, corporate services, or secure supply chain

A state in which a company is equally proficient at both innovation and cost proficiency, achievable only via digital technologies. Future-ready companies are typically agile, reuse modular capabilities, make their data a strategic asset, and partner to participate in ecosystems

Companies characterized by digital industrialization apply the best practices of automation to their operations. They use features that strengthen the company, and turn them into modular and standardized digitized services

Companies offering an integrated experience provide a better-than-industry-average customer experience despite having complex operations

Significant, disruptive changes that affect most of a company's customers, employees, and partners

When organizations resist change by holding on to existing approaches that have proven successful

One of four “explosions”—significant, disruptive changes—that a company must anticipate and manage during a digital business transformation. Surgery involves changes that remove organizational complexity and help the company focus on producing integrated customer offerings

A key business capability that is nurtured and enduring and used by multiple parts of the organization. It includes technology, governance, data, business processes, and APIs. Sometimes talent is included. Platforms can be used internally. They can also be opened up externally and sometimes commercialized.

Using existing platform infrastructure, components, or services across multiple applications to save development time and cost, and to improve time-to-market and performance

An enterprise that can respond immediately to opportunities and challenges by executing key business processes through automated digitized operations and employee-made data-driven decisions, supported by governance and risk guardrails.

When an enterprise transitions legacy applications to cloud platforms and digitizes its “crown jewels” (the components and capabilities it is best known for)

A complex landscape of business processes, systems, and fragmented data that supports an extensive catalog of products and services organized in enterprise silos

Digital partnering refers to increasing reach and range via digital connections with other companies

Companies increase partnering range by adding new products through digital partnering

Companies increase partnering reach by adding new customers through digital partnering

Partnering readiness is a set of capabilities that enables a company to become a strong participant in an ecosystem. It includes distinctiveness, openness, and organizational readiness

Partnering strength is a set of capabilities that enables a company to coordinate effectively with digital partners. It includes joint goals, sharing benefits, and sharing information

Small, nimble companies that apply digital technology to a particular industry or function

A set of capabilities to scale carbon emission reduction—the focus of MIT CISR’s research in the domain of sustainability—including green IT, green products, and green digital offerings

A digital sustainability capability focused on replacing carbon-intensive technologies and captures tracking and optimizing of technology emissions

A digital sustainability capability focused on decarbonizing products and production and captures tracking and optimizing product emissions

A digital sustainability capability focused on providing digital offerings that have lower carbon footprints or that help suppliers, partners, and customers track and manage their emissions

A strategic goal that is embedded into company strategy and guides the company’s focus on building digital sustainability capabilities