The new era of AI: From single agents to digital teams
Across industries, organizations are racing to harness the power of AI. The potential in intelligent automation alone motivated a $252.3 billion corporate AI investment in 2024,1 and those investments have been evolving almost as quickly as this rapidly changing technology itself.
While longer-standing AI technologies like machine learning and chatbots continue to perform well, agentic AI has moved to the front of the pack, offering the kind of autonomous decision-making that companies crave. Early wins with generative AI—drafting emails, summarizing documents, automating routine tasks—have shown what’s possible when a single intelligent agent is put to work.
Microsoft defines agentic AI as the pairing of traditional software strengths—such as workflows, state, and tool use—with the adaptive reasoning capabilities of large language models (LLMs). This allows agents to understand intent, take action, and interact with other systems dynamically, moving beyond the limits of rule-based automation.
What are the benefits of LLMs?
As organizations look to scale AI across more of their operations, many are finding that a single agent can’t always manage complex, multi-step tasks. This is where multi-agentic systems become valuable.
Multi-agentic systems use a series of agents, with a single coordinating agent, to work as a sort of AI team. The coordinating agent works to understand complex queries and delegate workflows to other agents, making multi-step, multi-system queries possible. In collaboration with people who are essential for escalation, understanding significant ambiguity, and creative thinking, multi-agentic systems are becoming integral to digital-first workforces.
Microsoft AI
Because these AI teams often operate across different tools and systems, organizations need solutions that are secure and enterprise-ready. With Microsoft technologies, agentic systems are built with the security, compliance, and reliability businesses expect. For a deeper dive on multi-agentic AI, read “Designing Multi-Agent Intelligence” on Microsoft Dev Blogs.
Today, Microsoft customers are already seeing the impact of multi-agentic AI. Here are three stories out of the Microsoft AI Co-Innovation Lab in San Francisco that show how multi-agentic AI is transforming security, science, and retail.
Three real-world examples of multi-agentic AI transformation
1. Contraforce: Turning the tide in cybersecurity
In cybersecurity, every second matters. For managed service providers (MSPs), responding to threats quickly can mean the difference between business as usual and a major incident. Contraforce, a Microsoft partner, set out to change the game with a multi-agentic security delivery platform built on Microsoft Foundry.
The multi-agentic solution automates 90% of incident investigations and response tasks, working as an always-on security operations team to analyze security data, identify suspicious activities, and autonomously managing incidents. These autonomous AI agents don’t just automate tasks—they help create a new cyber defense workforce.
The results are striking:
- Incident response times plummeted from 30 minutes to just 30 seconds.
- The cost per incident dropped from $15 to less than $1.
- MSPs can now scale their services without scaling their teams.
Contraforce’s story is a testament to how agentic AI can transform security operations from reactive to proactive, delivering speed, scale, and cost-efficiency.
3. Stemtology: Accelerating discovery in health sciences
Medical innovations can move slowly, and sometimes for good reason. But in regenerative medicine, lengthy research cycles and complex data analysis can be optimized with AI intervention.
Regenerative medicine innovator Stemtology worked with the Microsoft AI Co-Innovation Lab to accelerate biomedical discovery using a multi-agentic platform.
By combining Azure Cognitive Search, GPT-based agents, and domain-specific knowledge graphs, Stemtology’s system allows agents to:
- Parse scientific literature
- Generate therapeutic hypotheses
- Design and evaluate experiments
The impact? Research timelines have been cut by up to 50% at Stemtology. Minimum viable products are delivered in weeks instead of months. And the path from idea to patient-ready therapy is shorter than ever. This has freed up researchers to focus on highly complex evaluation and design strategies for treatments, rather than spending hours on gathering and synthesizing research.
Stemtology’s journey shows how agentic AI can support critical human discovery and bring life-saving treatments closer to reality.
3. SolidCommerce: Personalizing customer engagement at scale
For retailers, delivering personalized experiences while managing vast product catalogs and backend operations is a constant challenge. SolidCommerce specializes in providing AI solutions that address these challenges in the retail industry.
Hoping to address time-consuming support processes, inconsistent customer communications, and operational inefficiencies handling customer support, they approached the Microsoft AI Co-Innovation Lab in San Francisco to create an AI agent that could automate accurate and brand-aligned responses to meet customer needs.
Their solution brings together multiple agents for customer triage, FAQ handling, account management, product recommendations, and compliance checks. Built on Microsoft’s Agentic AI framework and integrated with Microsoft Copilot Studio and Foundry Agent Service, the system is easy to deploy and scale.
The payoff:
- Richer, multimodal customer experiences
- Scalable automation across channels
- Real-time personalization with memory and context
SolidCommerce’s story demonstrates how multi-agentic AI can turn retail complexity into seamless, intelligent engagement, ensuring customer satisfaction to keep pace with technological change.
Learn more about the agentic advantage
Microsoft customers are realizing benefits every day across industries. As we’ve seen in the customer examples above, multi-agentic AI delivers speed and scale in operations, accelerates innovation in research and development, and enables personalized engagement at scale.
Microsoft AI Co-Innovation Labs
Accelerate your AI projects with personalized help from our Microsoft Technology Experts
Multi-agentic AI isn’t just a technical upgrade—it’s a strategic shift. And companies that harness these systems to transform legacy processes can benefit not only from automation, but from truly intelligent optimization.
Learn how other customers are transforming with AI and explore creating your own generative AI proof of concept at Microsoft AI Co-Innovation Labs.
1 The 2025 AI Index Report, Stanford HAI.
