How will enterprise and SME organizations change after AGI?

Introduction

Today's AI systems (ChatGPT, Gemini, Claude and the like) can already be said to know more than the average specialist. Once that capability is generalized even further into AGI (Artificial General Intelligence), everyday life is expected to change dramatically.
In this article, we look at how the organizations of large corporations and small and medium-sized enterprises (SMEs) are likely to change.

References

There is a great deal of information circulating about AGI. Here, we have restricted ourselves to highly credible sources, such as those from the following types of authors:

• Leading AI developers: OpenAI, Google, etc.
• Leading consulting firms: McKinsey & Company, BCG, etc.
• Leading securities and investment firms: Goldman Sachs, Sequoia Capital, etc.
• Leading research institutions: Harvard University, Stanford University, etc.
• Leading academic journals: Nature, Science, etc.
• International organizations: World Economic Forum, OECD, etc.

Our analysis is based on the sources listed below. Please refer to them directly for further detail.

• Acemoglu, D., & Restrepo, P. (2018). Artificial intelligence, automation and work (NBER Working Paper No. 24196). National Bureau of Economic Research.
• AlixPartners. (2025). Farewell, SaaS: AI is the future of enterprise software.
• Altman, S. (2021). Moore's law for everything.
• Amodei, D. (2024). Machines of loving grace.
• Bai, Y., Kadavath, S., Kundu, S., Askell, A., Kernion, J., Jones, A., ... & Kaplan, J. (2022). Constitutional AI: Harmlessness from AI feedback. Anthropic. (arXiv:2212.08073)
• Appel, R., McCrory, P., Tamkin, A., McCain, M., Neylon, T., & Stern, M. (2025). The Anthropic Economic Index Report: Uneven geographic and enterprise AI adoption. Anthropic.
• Aschenbrenner, L. (2024). Situational awareness: The decade ahead.
• BCG. (2025). Software engineering, UI/UX, product management, quality assurance, data science: How tasks, talent, and teams will change [Presentation slides].
• Bengio, Y., et al. (2025). The international scientific report on the safety of advanced AI.
• Bostrom, N. (2014). Superintelligence: Paths, dangers, strategies. Oxford University Press.
• Bremmer, I., & Suleyman, M. (2023). The AI power paradox: Can states learn to govern artificial intelligence—before it’s too late? Foreign Affairs.
• Deloitte. (2026). SaaS meets AI agents: Transforming budgets, customer experience, and workforce dynamics. Deloitte Center for Technology, Media & Telecommunications.
• Patel, D. (Host). (2025). Interview with Satya Nadella [Audio podcast episode]. Dwarkesh Patel Podcast.
• Eloundou, T., Manning, S., Mishkin, P., & Rock, D. (2023). GPTs are GPTs: An early look at the labor market impact potential of large language models.
• Gawdat, M. (2021). Scary smart.
• Goldman Sachs. (2026). How will AI affect the US labor market?
• IBM Institute for Business Value. (2025). AI-fueled operations. IBM.
• Lykkegaard, B. (2026). Is SaaS dead? Rethinking the future of software in the age of AI. IDC Tech Buyer.
• Cazzaniga, M., Jaumotte, F., Li, L., Melina, G., Panton, A. J., Pizzinelli, C., Rockall, E., & Tavares, M. M. (2024). Gen-AI: Artificial intelligence and the future of work (IMF Staff Discussion Note SDN/2024/001). International Monetary Fund.
• International Telecommunication Union (ITU). (2025). The annual AI governance report 2025: Steering the future of AI.
• Kissinger, H., Schmidt, E., & Huttenlocher, D. (2021). The age of AI: And our human future. Little, Brown and Company.
• Korinek, A. (2024). Economic policy challenges for the age of AI (NBER Working Paper No. 32980). National Bureau of Economic Research.
• Lu, C., Lu, C., Lange, R. T., Foerster, J., Clune, J., & Ha, D. (2024). The AI Scientist: Towards fully automated open-ended scientific discovery. arXiv preprint arXiv:2408.06292.
• Maslej, N., Fattorini, L., Perrault, R., Gil, Y., Parli, V., Kariuki, N., ... & Oak, S. (2025). The AI index 2025 annual report. Stanford Institute for Human-Centered Artificial Intelligence.
• Maslej, N., Fattorini, L., Perrault, R., Gil, Y., Parli, V., Kariuki, N., ... & Oak, S. (2025). The AI index 2025 annual report. Stanford Institute for Human-Centered Artificial Intelligence.
• McKinsey & Company. (2025). The agentic organization: Contours of the next paradigm for the AI era.
• McKinsey & Company. (2026). The state of organizations 2026.
• McKinsey Global Institute. (2023). The economic potential of generative AI: The next productivity frontier.
• Microsoft. (2025). Work trend index annual report 2025: The year the frontier firm is born.
• OpenAI. (2023). Governance of superintelligence.
• Ord, T. (2020). The precipice: Existential risk and the future of humanity. Bloomsbury.
• PwC. (2025). The fearless future: PwC’s 2025 global AI jobs barometer.
• Pavel, B., Ke, I., Smith, G., Brown-Heidenreich, S., Sabbag, L., Acharya, A., & Mahmood, Y. (2024). How artificial general intelligence could affect the rise and fall of nations: Visions for potential AGI futures (RAND Corporation Research Report PE-A3776-1). RAND Corporation.
• Russell, S. J. (2019). Human compatible: Artificial intelligence and the problem of control. Penguin Books.
• Suleyman, M. (2023). The coming wave: Technology, power, and the twenty-first century's greatest dilemma. Crown.
• Susskind, D. (2020). A world without work: Technology, automation, and how we should respond. Metropolitan Books.
• Tegmark, M. (2017). Life 3.0: Being human in the age of artificial intelligence. Knopf.
• Toner-Rodgers, A. (2024). Artificial intelligence, scientific discovery, and product innovation. arXiv preprint arXiv:2412.17866.
• Bengio, Y., Clare, S., Prunkl, C., Murray, M., Andriushchenko, M., Bucknall, B., ... & Mindermann, S. (2026). International AI safety report 2026 (DSIT 2026/001). Department for Science, Innovation and Technology.
• Bengio, Y., Clare, S., Prunkl, C., Murray, M., Andriushchenko, M., Bucknall, B., ... & Mindermann, S. (2026). International AI safety report 2026 (DSIT 2026/001). Department for Science, Innovation and Technology.
• World Economic Forum. (2025). Future of jobs report 2025.
• Yamada, Y., Lange, R. T., Lu, C., Hu, S., Lu, C., Foerster, J., Clune, J., & Ha, D. (2025). THE AI SCIENTIST-V2: Workshop-level automated scientific discovery via agentic tree search. arXiv preprint arXiv:2504.08066.

Organizational Structure

1. A Fundamental Shift in Structure: From Hierarchies to "Agentic Networks"

• From the "Org Chart" to the "Work Chart": The traditional pyramid-shaped org chart, siloed by function and department, will break down. In its place will come a dynamic "work chart" organized around the jobs to be done and the business outcomes to be achieved.
• Agile "Fusion Teams": For each project, humans and AI agents will come together, each contributing the expertise required, to form cross-functional, ultra-flat hybrid teams (pods or fusion teams). These teams disband once the task is complete, making highly fluid, agile organizational structures the norm.
• Fewer Middle Managers and Ultra-Flat Organizations: As AI agents autonomously take over routine tasks such as cross-departmental coordination, information relay and progress tracking, the need for the middle-management layer that used to handle this coordination will shrink, and organizations will flatten out.
• Scaling With Very Small Teams: A team of just two to five people, supervising a fleet of 50 to 100 specialized AI agents (an "agent factory"), will be able to handle work such as customer onboarding or financial close end to end, at a scale never before possible.

2. Talent Models and the Emergence of New Roles and Departments

• Every Employee Becomes an "Agent Boss": From new hires to the executive suite, everyone will direct and manage one or more AI agents, taking on the role of an "agent boss" who uses them to amplify their own impact.
• Three New Talent Archetypes: In an agentic organization where AI handles execution, the following three talent models will matter more than functional specialization.
  • M-shaped general managers: People with broad command of many domains who design, supervise and optimize entire human-AI hybrid workflows.
  • T-shaped deep specialists: People with deep expertise in a specific domain who handle the exceptions AI agents cannot and guarantee quality.
  • AI-empowered frontline: People who, with AI support running in the background, specialize in human empathy and sophisticated interpersonal communication (sales, HR, healthcare and so on).
• A New "Intelligence Resources" Department: To manage the optimal allocation and coordination of digital and human labor centrally at the organizational level, the traditional IT and HR departments will merge into a new "Intelligence Resources" department, which will become a source of competitive advantage. Moderna's decision to unify HR and IT leadership is an early example of this.
• Shifting Skill Requirements: The value of "execution" skills such as coding and data entry will decline, while systems thinking, problem framing, the ability to properly verify AI output, ethical judgment and human interpersonal skills (EQ) will become overwhelmingly important.

3. The Evolution of Operations and Governance (Transformation in Large Corporations)

• Shared Services Evolve Into "AI-Native GBS": In large corporations, the physical shared services centers (SSCs) originally set up to cut costs and streamline processes will evolve into virtual "Global Business Services" (GBS) built around AI agents. Here, AI executes processes end to end, and the GBS becomes a central hub that drives company-wide innovation under human oversight.
• Built-In Real-Time Governance and Guardrails: With AI agents operating autonomously around the clock, traditional manual audits on an annual or monthly cycle will no longer suffice. Instead, distributed governance will become essential: "compliance agents" and "guardrail agents" that monitor the executing agents will be embedded directly into systems, enforcing data privacy, financial standards and the like in real time.
• Optimizing the "Human-Agent Ratio": For each business process, companies will need to optimize a new management metric, the human-agent ratio, which captures the balance of how many humans should supervise how many AI agents for maximum efficiency and minimum risk.

4. The Collapse of the Scale Barrier (SMEs Become "Frontier Firms")

• Scalability Through Intelligence on Demand: The advanced expertise and scalability that were once the preserve of large corporations will become available to SMEs through AI, on demand and at extremely low cost, like water from a tap ("intelligence on tap").
• The Rise of the Frontier Firm: A wave of "frontier firms" will emerge: small teams that borrow the labor of AI agents to match the speed and output of large corporations. As a result, the "scale-based competitive barrier" that has long separated large companies from SMEs will largely collapse.

Talent Models, Roles and Departments

1. A Radical Change in the Talent Model (From Execution to Supervision and Orchestration)

• From "Executing Tasks" to "Supervising Outcomes": The employee's role will shift from personally executing tasks to defining goals, making trade-off decisions and supervising the workflows that AI agents execute. Performance evaluation will change fundamentally as well, from how well tasks were completed to how skillfully the employee orchestrated AI agents and created value.
• Three New Talent Archetypes in Demand: In an organization where AI handles task execution, the following three talent models, distinct from traditional functional specialization, will become important.
  1. M-shaped general managers: People with knowledge spanning a wide range of domains who design, supervise and optimize entire human-AI hybrid workflows.
  2. T-shaped deep specialists: People with extremely deep expertise in a specific domain who handle the exceptions AI agents cannot, assure quality and guide the models.
  3. AI-empowered frontline: People who, with AI support running in the background, specialize in human empathy and sophisticated interpersonal communication (sales, HR, healthcare and so on).
• Dramatic Shifts in Skill Requirements: The value of "execution" skills such as coding and data entry will decline, while systems thinking, critical thinking, problem framing, exception handling and uniquely human soft skills and ethical judgment will become overwhelmingly important.

2. New Roles Emerge and Existing Roles Are Redefined

• Every Employee Becomes an "Agent Boss": From new hires to the CEO, every employee will direct and manage their own dedicated AI agent (or several) to scale their work, taking on the role of an "agent boss."
• The Rise of AI-Native Roles: New positions such as "AI Trust & Safety Lead" and "AI Product Owner" will emerge to manage AI systems. Roles premised on human-AI collaboration, such as "LLM (large language model) Product Manager," "Prompt Ops" and "Agent Quality Assurance," will also proliferate rapidly within organizations.
• Upgrading Existing Roles: Rather than simply disappearing, existing jobs will evolve into higher-value roles. Data entry clerks, for instance, will move up to become data analysts, while paralegals will shift toward verifying the output of AI tools and focusing on critical thinking and collaboration with clients.
• The Changing Role of Middle Management: As AI agents autonomously handle routine tasks such as coordination, information relay and progress tracking, the need for middle managers will diminish and organizations will become extremely flat. The managers who remain will focus on what AI cannot replace: coaching their teams, people management and higher-level strategic judgment.

3. Departmental Restructuring and the Fusion of Organizational Functions

• Creation of an "Intelligence Resources" (IR) Department: To manage the optimal allocation and coordination of human and digital (AI) labor centrally across the company, the traditional IT and HR departments will merge into a new "Intelligence Resources" department that serves as a source of competitive advantage. Just as Moderna unified its HR and IT leadership, AI will come to be positioned not as a mere IT system but as a force that shapes the workforce itself.
• The Collapse of Functional Silos and the Formation of "Fusion Teams": The walls between siloed departments (sales, development, HR and so on) will come down, and organizations will move toward a "work chart" (a workflow map) organized around the jobs to be done and the outcomes to be achieved. Cross-functional "fusion teams" and agile "pods" that blend technology, data, HR and business specialists will be assembled and then disbanded once the task is complete, making this kind of fluid organizational operation the norm.
• Shared services become "AI-native Global Business Services (GBS)": The physical shared service centers (SSCs) that large enterprises set up to cut costs and streamline processes have hit their limits. Going forward, they will evolve into virtual "Global Business Services (GBS)" built around AI agents that automate processes end to end, becoming the central hub that drives innovation across the entire company.

4. How the impact differs between large enterprises and SMEs

• Large enterprises (overcoming legacy and managing large-scale change): Large enterprises face the challenge of breaking through rigid existing organizational structures and legacy systems. Success will hinge on carrying out large-scale reskilling for the entire workforce and building embedded governance frameworks—such as deploying compliance agents—that monitor the decisions of AI agents in real time.
• SMEs (the leap to becoming "Frontier Firms"): By tapping "intelligence on tap"—intelligence available on demand—SMEs can acquire, at extremely low cost, the advanced expertise and workforce scale that only large enterprises could previously command. With the rise of "Frontier Firms," in which small teams direct vast fleets of AI agents to generate enormous value, the competitive barriers imposed by company size will largely crumble.

Operations and Governance

1. Redefining operations: from hierarchies and silos to "agent-driven flows"

• Shared services become "AI-native GBS": The physical "shared service centers (SSCs)" that large enterprises built to reduce costs will evolve into virtual "Global Business Services (GBS)" with AI agents at their core. Humans and AI will work together to automate processes end to end, and these units will function as the central hub driving innovation across the entire enterprise.
• From the org chart to the "work chart": The basic unit of operations will shift from the siloed org chart of "departments" and "hierarchies" to "workflows" organized around tasks and outcomes. Humans will step away from executing processes by hand and instead focus on orchestrating—directing and supervising—fleets of AI agents that collaborate autonomously across multiple departments.
• Real-time, autonomous execution of work: Operations in every domain—finance, supply chain, physical asset management—will become autonomous. For example, AI agents will autonomously schedule predictive maintenance based on historical data, head off supply chain bottlenecks using real-time market data, and even propose budgets and scenario forecasts on the spot, a practice known as "agentic budgeting" that will become widespread.

2. The evolution of governance: real-time control and "agents watching agents"

• From "after-the-fact audits" to "real-time embedded governance": In a world where AI agents operate autonomously around the clock, the manual audits and governance conducted on a quarterly or monthly basis will simply be too slow to function. To keep pace, companies will be compelled to build real-time, data-driven governance directly into their business processes and systems themselves (embedded governance).
• Agents monitoring agents (guardrail agents): To ensure compliance and avert brand risk, companies will deploy "policy agents" and "compliance agents" that monitor and audit the behavior of the AI agents doing the actual work. This creates "guardrails" in which AI itself checks in real time for budget overruns, data privacy violations, and ethical deviations—and blocks them.
• Redefining human accountability: Even when the bulk of operations is completed with humans out of the loop, legal accountability and responsibility for final decisions will remain with humans. The role of executives and compliance officers will be elevated from approving individual tasks to "designing the rules and policies that AI systems must follow" and "monitoring and managing outliers."

3. How the impact and strategies differ by company size (large enterprises vs. SMEs)

• "Hyper-scaling" for SMEs and startups (Frontier Firms): With the advent of AGI, advanced intelligence becomes available on demand—"intelligence on tap," as readily as tap water. This lets SMEs acquire, at extremely low cost, the advanced expertise and system scalability that were once the exclusive preserve of large enterprises. A new type of company will emerge—the "Frontier Firm," composed of a handful of people and dozens of specialized AI agents—that delivers speed and productivity rivaling large enterprises despite its small headcount, dramatically reshaping the competitive landscape.
• The challenge for large enterprises: dismantling legacy and adopting a distributed approach: In large enterprises, massive legacy systems accumulated over many years (IT monoliths) and rigid, siloed structures will be the biggest obstacles to deploying AI agents seamlessly. To become agile, large enterprises need to adopt a "distributed approach"—in which a central specialist team manages AI models and governance standards while boldly delegating authority over AI use to business units on the front lines—striking a balance between flexibility and control.

Impact on Large Enterprises and SMEs

1. The overwhelming advantage of large enterprises in foundation model development and the drift toward oligopoly

• Astronomical development costs and "winner-takes-all" dynamics: Developing and operating cutting-edge AGI and foundation models requires gigantic data centers costing hundreds of billions of dollars, vast computing resources (such as state-of-the-art GPU clusters), and extensive proprietary datasets. Because of these enormous capital requirements, the race to develop AGI will concentrate in a small number of deep-pocketed tech giants, greatly intensifying market concentration and "winner-takes-all" tendencies.
• Barriers to entry for startups and deepening dependence: We have entered a phase in which it is no longer possible for a lone startup to build ultra-advanced AGI (superintelligence) from scratch on its own. Startups are therefore forced to rely on the computing infrastructure and models provided by large enterprises. In exchange for access to frontier technology, they end up accepting acquisitions by or partnerships with those large players—a dynamic that further entrenches the power of the tech giants.

2. SMEs and startups become "Frontier Firms" as the barriers of scale collapse

• Gaining scalability through "intelligence on demand": While large enterprises monopolize the development of foundation models themselves, SMEs and startups will be able to access world-class intelligence cheaply through APIs and high-performing open-source (open-weight) models. As a result, the "advanced expertise" and "operational scalability" once monopolized by large enterprises become available at extremely low cost even to companies with limited financial resources.
• The rise of the "Frontier Firm" and agility as a weapon: New organizational forms will emerge—called the "Agentic Enterprise" or the "Frontier Firm"—in which a handful of people work alongside dozens to hundreds of specialized AI agents. Leveraging their lean structure and agility, startups can be first to build AI-native workflows, achieving speed and results on par with large enterprises despite their small size, and unleashing disruption that upends existing business models.

3. The challenges facing large enterprises (overcoming legacy and change management)

• Breaking through legacy systems and organizational silos: Large enterprises hold the advantage in capital and customer base, but legacy systems (aging IT infrastructure) built up over many years and rigid, siloed departmental structures become serious shackles when it comes to deploying AI agents seamlessly. To maximize the benefits of AI, they cannot simply automate parts of existing processes; they must rebuild entire workflows from scratch with AI agents as the starting premise—a shift "from structure to flow."
• Committing to large-scale change management and reskilling: For large enterprises, the greatest challenge lies not in adopting the technology itself but in transforming leadership and organizational culture. Whether they can dispel employees' anxiety about and resistance to being replaced by AI, carry out large-scale reskilling premised on working alongside AI agents, and dynamically reallocate budgets and talent toward strategic priorities will be the key to sustaining their competitive advantage.

4. A shift in the sources of competitive advantage (proprietary data and optimization)

• The rising value of proprietary data: As high-performance AI models themselves become ubiquitous and commoditized, the differentiator between companies shifts from "which AI you use" to "how well you accumulate your own proprietary data and have AI learn from and leverage it in a secure environment." Large enterprises hold an overwhelming advantage in the vast stores of customer and operational data they have amassed, but SMEs can build their own distinctive competitive edge by specializing in deep, domain-specific data within particular niche markets.

Closing Thoughts

The advance of AI will demand sweeping transformation from every organization, regardless of whether it is a large enterprise or an SME. Organizational leaders will need to make decisions that are both careful and swift—including some hard ones.