MathWorks has introduced Release 2026a (R2026a) of MATLAB and Simulink, integrating AI copilots and agentic workflows directly into engineering environments to accelerate embedded systems development while maintaining rigour and traceability.
At the core of the release are Simulink Copilot and Polyspace Copilot, designed to bring context-aware, “grounded AI” assistance into model-based design and software verification workflows. Unlike generic AI tools, these copilots are trained on engineering models, project workflows, and domain-specific documentation, ensuring outputs remain aligned with strict engineering standards.
Simulink Copilot enables engineers to interpret model behaviour, generate explanations, identify issues, and suggest next steps, significantly improving productivity in complex system design. By guiding users through structured workflows, it also promotes consistency and repeatability across development and verification processes.
On the software side, Polyspace Copilot enhances static code analysis and defect resolution, helping engineers understand vulnerabilities and fix issues earlier in the development lifecycle. Complementing this, Polyspace as You Code introduces real-time validation of C and C++ code—including AI-generated code—bringing shift-left testing and continuous quality assurance into embedded software development.
Beyond copilots, MathWorks is extending AI into agentic engineering workflows through the MATLAB MCP Core Server and MATLAB Agentic Toolkit. These capabilities allow teams to integrate MATLAB and Simulink into automated, multi-step engineering processes, enabling more efficient transitions from design to verification and production.
The release also introduces enhancements across simulation, testing, and integration, including improved Python interoperability, digital twin modelling via Functional Mockup Units (FMUs), and advanced signal processing and geospatial analysis tools. Together, these updates strengthen MATLAB and Simulink as a unified platform for end-to-end system engineering.
According to Avinash Nehemiah, the focus is on delivering AI capabilities that enhance productivity without compromising engineering discipline, ensuring that speed, accuracy, and trust remain central to system development.
Overall, R2026a reflects a broader shift in engineering software—from standalone tools to AI-augmented, model-driven development platforms, where copilots and agentic systems enable faster innovation while preserving the precision and reliability required for complex, safety-critical systems.