The enterprise AI conversation is changing.
A year ago, many IT leaders were asking whether they should deploy an AI agent. Today, a more consequential question is emerging. How should organizations coordinate multiple agents working across departments, applications and business processes?
Interest in this model is accelerating. Gartner reported a 1,445 per cent increase in enquiries about multi-agent systems between the first quarter of 2024 and the second quarter of 2025.¹ Gartner also predicts that 40 per cent of enterprise applications will include task-specific AI agents by the end of 2026, up from less than 5 per cent in 2025.²
The architecture supporting this shift is multi-agent orchestration. It is becoming an important consideration for CIOs, CTOs and enterprise architects seeking to move beyond isolated AI experiments.
From solo performers to coordinated networks
A single AI agent can be effective when given a clearly defined task. It can interpret a request, use an approved tool and return a result.
Most enterprise workflows, however, are not isolated.
They cross system boundaries, involve different areas of expertise and depend on decisions made throughout a broader process. A customer request might require an agent to retrieve account information, another to assess contractual obligations and a third to initiate an operational workflow.
A multi-agent architecture divides this work among specialised agents with defined responsibilities. An orchestration layer coordinates how tasks are assigned, how context is shared and when work is transferred between agents.
The result is less like a single digital assistant and more like a coordinated team. Each agent performs a defined role while the wider system manages the workflow.
This division of responsibility can improve modularity and make individual components easier to test, govern and replace. It also introduces additional complexity. Microsoft’s current guidance recommends starting with a single agent and separating responsibilities only when modularity, security boundaries or specialist capabilities make a multi-agent design necessary.
Five orchestration patterns that matter
Several recurring patterns are shaping the way multi-agent systems are designed.
Sequential Orchestration
One agent completes a task before passing its output to the next agent. This approach is suitable for structured, linear processes such as document verification, invoice handling and compliance reviews.
Its strength is predictability. Each stage can be recorded and evaluated independently. Its weakness is that an error or delay early in the sequence can affect every later stage.
Parallel Orchestration
Multiple agents work at the same time on separate parts of a problem. Their results are then combined by an orchestrator or aggregation agent.
This pattern can support research, security analysis, market assessment and risk evaluation. It reduces total processing time when tasks do not depend on one another, although conflicting results still require a method of reconciliation.
Hierarchical Orchestration
A supervisor or manager agent breaks a larger objective into smaller tasks and assigns them to specialist agents.
This approach can make complex workflows easier to manage, but it places significant responsibility on the supervisor. Poor planning or task allocation at the top can reduce the quality of the entire system.
Handoff Orchestration
An initial agent assesses a request and transfers it to the most appropriate specialist.
This model resembles a service desk or contact center. It can be effective when requests fall into recognisable categories, but routing rules and escalation paths must be carefully designed.
Iterative Orchestration
An agent or group of agents repeatedly reviews and improves an output until it meets defined acceptance criteria.
This pattern is useful for drafting, software testing, data validation and quality assurance. Clear stopping conditions are essential to prevent unnecessary processing, rising costs or endless revision loops.
These patterns are not mutually exclusive. A single enterprise workflow might use a routing agent at the entry point, a supervisor within a business domain and sequential or parallel processing inside individual tasks.
The real architectural skill lies in combining these approaches without creating unnecessary complexity.
Why interoperability matters
One of the most significant developments in multi-agent architecture has been the emergence of open communication standards.
Google introduced the Agent2Agent protocol, commonly called A2A, in April 2025. The protocol was designed to allow agents created with different frameworks or by different vendors to discover capabilities, exchange information and delegate tasks.
Google transferred the A2A project to the Linux Foundation in June 2025, establishing vendor-neutral governance for the standard. By April 2026, the Linux Foundation reported that more than 150 organizations supported A2A, with integrations across major cloud platforms and early production use.³
This matters because most large organizations do not operate within a single technology ecosystem.
An enterprise may use Microsoft for productivity and workflow automation, Salesforce for customer relationship management, Google Cloud or Amazon Web Services for infrastructure, and custom-built agents for industry-specific functions.
Without an interoperability standard, every connection between those systems could require a bespoke integration. A2A provides a common contract through which compatible agents can communicate while keeping their internal implementations private.
A2A does not eliminate integration, security or governance work. Organizations still need to manage identity, authentication, authorisation and data exposure. It can, however, reduce the amount of custom communication logic required when connecting compatible agent platforms.
Microsoft is moving in the same direction. Copilot Studio now supports multi-agent orchestration and connections to external agents using the A2A protocol. Microsoft has also made computer-using agents generally available, allowing approved agents to interact with websites and desktop applications when suitable application programming interfaces are unavailable.
The direction across the major platforms is becoming clearer. Enterprise agents are increasingly being designed as parts of coordinated systems rather than disconnected point solutions.
What multi-agent systems mean for IT architecture
Deploying a multi-agent environment is not simply a matter of creating more agents.
Each additional agent introduces dependencies, permissions, context requirements and possible failure points. The overall system must therefore be designed around several architectural priorities.
State and context management
When one agent transfers work to another, the receiving agent needs enough information to continue the task accurately.
Passing too little context can produce incomplete or inconsistent outcomes. Passing too much can increase cost, expose unnecessary information or overwhelm the receiving agent.
Architects need to define what information moves between agents, where shared state is stored, how long it is retained and which agents are permitted to access it.
Checkpointing can also help a system resume from a known stage after an error rather than restarting the entire workflow.
Orchestration and control
A clear control mechanism is needed to assign tasks, manage dependencies and determine what happens when an agent fails.
This does not always require a single, central orchestrator. Some environments may use hierarchical or distributed coordination. Regardless of the model, the organization needs an identifiable control plane for policy enforcement, access management and auditability.
Without one, agents can duplicate work, issue contradictory actions or operate beyond their intended responsibilities.
Observability
Traditional application monitoring focuses on infrastructure health, availability, response times and errors.
Agent observability must examine a broader set of questions.
Did the system select the appropriate agent? Was the necessary context transferred? Which tools and data sources were used? Why was a particular action taken? Did the workflow remain within its cost and risk limits?
Useful observability should capture agent interactions, tool calls, decisions, exceptions, latency, token usage and cost. It should also make the overall workflow understandable to technical teams and, where required, auditors or business owners.
Identity and access
Each agent should have its own defined identity and permissions.
Using shared or overly broad credentials makes it difficult to determine which agent performed an action and increases the potential impact of a compromised component.
Access should follow least-privilege principles. An agent should only be able to retrieve or modify the information needed for its assigned function.
Governance cannot be an afterthought
As agents gain greater autonomy, governance becomes part of the technical architecture rather than a separate compliance activity.
A minimum governance model should include:
- Role-based access controls defining what each agent can view, create or modify
- Tamper-resistant audit records capturing significant actions and decisions
- Human review requirements for sensitive, high-impact or irreversible decisions
- Data classification and retention rules covering information shared between agents
- Testing for reliability, security, bias and unexpected behaviour
- Anomaly detection and mechanisms for suspending individual agents
- Version control and change-management processes for prompts, models, tools and policies
- Clear accountability for each agent and the business process it supports
Human involvement should be based on risk rather than added indiscriminately.
Routine, reversible and low-risk actions may be suitable for higher levels of automation. Decisions affecting employment, financial commitments, regulated information, customer rights or critical systems generally require stronger controls and human oversight.
The business opportunity
The economic potential surrounding generative AI is substantial, although it should not be attributed exclusively to AI agents.
McKinsey estimates that generative AI could create between US$2.6 trillion and US$4.4 trillion in annual economic value across 63 use cases.⁴ Agents may help organizations capture part of that opportunity by coordinating tools, models and workflows, but the estimate applies to generative AI more broadly.
McKinsey’s 2025 global survey also found that AI high performers were more than three times as likely as other organizations to intend to use AI for transformative business change. More than one-third of these high performers reported allocating over 20 per cent of their digital budgets to AI technologies.⁵
The same survey demonstrates that scaled agent adoption remains relatively early. No more than 10 per cent of respondents reported scaling agents in any individual business function.⁵
This distinction is important.
Multi-agent architecture is progressing quickly, but most enterprises are still developing the operating models, controls and technical foundations needed to use it at scale. Organizations should avoid treating agent deployment as proof of business value. Success depends on workflow redesign, reliable data, measurable outcomes and appropriate governance.
The opportunity in front of IT leaders
Multi-agent orchestration is moving from research and experimentation into enterprise platforms and production environments.
That does not mean every process needs multiple agents. A single agent or conventional automation may remain the simplest and most reliable option for many tasks.
Multi-agent architecture becomes valuable when a workflow genuinely requires separate expertise, security boundaries, systems or decision stages.
The organizations best positioned to benefit will be those that design around clear business outcomes and establish the right foundations early. These foundations include interoperability, identity, observability, state management, cost controls and governance.
For IT leaders mapping an AI architecture, evaluating platforms or moving beyond isolated pilots, now is the time to determine where multi-agent systems add genuine value and where they add avoidable complexity.
Explore Insentra’s AI practice at AI Momentum to learn how we help organizations architect, govern and scale enterprise AI deployments.
Sources
¹ Gartner, Multiagent Systems in Enterprise AI, reporting a 1,445 per cent increase in multi-agent-system enquiries from Q1 2024 to Q2 2025.
² Gartner, Gartner Predicts 40% of Enterprise Apps Will Feature Task-Specific AI Agents by 2026, 26 August 2025.
³ Linux Foundation, A2A Protocol Surpasses 150 Organizations, Lands in Major Cloud Platforms, and Sees Enterprise Production Use in First Year, 9 April 2026.
⁴ McKinsey Global Institute, The Economic Potential of Generative AI, June 2023.
⁵ McKinsey and Company, The State of AI Global Survey 2025, November 2025.
