Microsoft has unveiled Majorana 2, a second-generation topological quantum computing chip developed using its agentic artificial intelligence platform, Microsoft Discovery, on June 9, 2026. The prototype chip features 12 topological qubits as a scalable model, with Microsoft aiming to build commercially viable quantum systems by 2029. Alongside this hardware advancement, the company introduced Scout, an autonomous assistant designed to manage workspace scheduling, emails, and meetings within Microsoft 365.
Understanding Majorana 2: Microsoft’s Topological Quantum Chip
Topological quantum computing is a highly specialized approach to building fault-tolerant quantum computers. Traditional quantum processors use qubits made of superconducting loops or trapped ions, which are highly sensitive to external environmental noise. In contrast, topological quantum systems store and process information non-locally by braiding quasiparticles, making the system inherently resistant to local environmental disturbances.
The Technology Behind Topological Qubits
The foundation of Microsoft’s quantum architecture lies in Majorana zero modes, which are emergent zero-energy states found at the boundaries of topological superconductors. Named after the Italian physicist Ettore Majorana, who proposed the existence of particles that act as their own antiparticles in 1937, these zero modes serve as the building blocks for topological qubits. By storing quantum information across a physical structure rather than at a single point, these qubits are protected from errors caused by thermal or electromagnetic fluctuations.
Key Specifications and Materials Stack
The newly introduced Majorana 2 prototype processor features 12 topological qubits configured to demonstrate scalability. Microsoft claims that the Majorana 2 processor is 1,000 times more reliable than its predecessor, Majorana 1, which was released in February 2025. A key performance metric is the qubit coherence lifetime, which now averages 20 seconds with certain configurations maintaining states for up to a minute. This represents a substantial improvement over the millisecond-scale lifetimes typical of traditional superconducting qubits.
To achieve this level of stability, the chip utilizes a completely redesigned materials stack. Specifically, engineers replaced the aluminum in the superconducting layers with lead and paired it with a semiconductor core made of indium arsenide and indium arsenide antimonide nanowires. This material combination widens the topological gap, providing stronger protection for the quantum states against external disturbances.
The Role of Microsoft Discovery in Quantum R&D
The development of the Majorana 2 chip was accelerated through the use of Microsoft Discovery, an Azure-based agentic artificial intelligence platform tailored for scientific and engineering research. Rather than relying solely on traditional trial-and-error laboratory experiments, Microsoft utilized AI agents to model, simulate, and optimize the hardware design.
The AI platform performed several critical functions during the development cycle:
- It optimized hundreds of physical and chemical design parameters for the new semiconductor nanowire stack.
- It automated complex cleanroom fabrication recipes to ensure structural precision at the nanoscale.
- It analyzed vast arrays of experimental test data to detect signatures of Majorana zero modes.
By automating these workflows, Microsoft Discovery reduced the quantum development lifecycle. The integration of agentic AI has enabled Microsoft to revise its roadmap, advancing the projected launch of a commercially viable, fault-tolerant quantum system from the original estimate of 2033 to a new target of 2029.
Microsoft Scout: Autonomous Work Agent for Microsoft 365
Alongside its quantum computing hardware milestone, Microsoft introduced Scout, a next-generation, autonomous personal assistant integrated into the Microsoft 365 software suite. Unlike traditional generative AI assistants that operate reactively based on direct user prompts, Scout functions as an active agent that runs continuously in the background to handle daily workplace operations.
The system is designed to carry out complex, multi-step actions across various applications such as Outlook, Teams, and OneDrive. Its core capabilities include:
- Autonomously managing, sorting, and drafting responses to email communications.
- Coordinating and scheduling meetings by analyzing calendar availability across multiple internal and external stakeholders.
- Preparing briefs, summarizing documents, and organizing files in advance of scheduled events without user intervention.
Scout is built on advanced developer frameworks, including the open-source OpenClaw and WorkIQ systems, which enable agents to understand user intent and execute tasks across application boundaries securely. This autonomous model represents a shift from static chat interfaces toward active, goal-oriented software agents.
Implications for the Quantum Computing Landscape
Microsoft’s push for a topological quantum computer represents a different approach than those taken by competitors like IBM and Google, which focus on superconducting transmon qubits. The main barrier to practical quantum computing is the high rate of errors caused by noise, a challenge known as decoherence. In superconducting or trapped-ion systems, researchers must implement complex Quantum Error Correction protocols, which require thousands of physical qubits to create a single, error-free logical qubit.
| Quantum Architecture | Qubit Type | Primary Error Protection Method | Scalability Challenge |
|---|---|---|---|
| Superconducting Loops | Local transmon qubits | Active Quantum Error Correction (large physical qubit overhead) | Sensitive to thermal noise and wiring complexity |
| Trapped Ions | Atomic qubits | Active error correction via laser controls | Scaling physical trap size and speed |
| Topological Systems | Majorana zero modes | Physical protection via non-local braiding | Material science fabrication and MZM verification |
Because topological qubits physically protect information at the hardware level, they require far less error-correction overhead. If Microsoft succeeds in demonstrating stable braiding with Majorana 2, it could scale quantum processors much faster. However, the scientific community continues to emphasize the need for independent, peer-reviewed verification, as isolating and proving the existence of Majorana zero modes is a complex experimental challenge.
Key Takeaways
- Microsoft has unveiled Majorana 2, its second-generation topological quantum computing chip featuring 12 topological qubits.
- The Majorana 2 processor is built using a redesigned superconducting materials stack that replaces aluminum with lead.
- Development of the chip was accelerated by Microsoft Discovery, an Azure-based agentic AI platform designed for scientific and engineering research.
- Microsoft has moved up its target for building a commercially viable, fault-tolerant quantum computer to 2029.
- The software agent Scout was launched to autonomously manage schedules, emails, and meetings within Microsoft 365 using the OpenClaw and WorkIQ frameworks.
