MathWorks adds AI copilot tools to MATLAB & Simulink

MathWorks has released R2026a of MATLAB and Simulink, adding new artificial intelligence tools for embedded systems development, including Simulink Copilot and Polyspace Copilot.

The release also updates MATLAB, Simulink and the Polyspace range as MathWorks expands the use of generative AI in engineering design and software verification.

Simulink Copilot is aimed at teams using Model-Based Design. Grounded in users' models, team processes and MathWorks documentation, it can generate model explanations, answer questions about model behaviour, and help users find relevant blocks and subsystems.

It is designed to help engineers isolate issues, suggest remedies and guide next steps during design work. It can also support standardised tasks to encourage more consistent development and verification practices.

Polyspace Copilot focuses on embedded software code analysis. It provides guidance based on Polyspace analysis results to help engineers interpret static analysis findings, understand issues and resolve them more efficiently.

R2026a also introduces Polyspace as You Code, which lets developers check C and C++ coding rules and identify defects and vulnerabilities as code is written, including code produced with AI-assisted tools. The addition reflects growing concern among engineering teams about how to verify software generated with machine assistance.

Verification focus

The wider Polyspace family also gains three additions: a new desktop application for unified configuration and results management; extensions to Polyspace Bug Finder with custom checkers and coding standards; and software-sanitising functions in Polyspace Test for dynamic analysis of runtime errors.

MathWorks' broader strategy has two strands. One is to embed copilots in the software environments already used by engineering teams, including MATLAB Copilot, Simulink Copilot and Polyspace Copilot. The other is to connect MATLAB and Simulink functions into agentic workflows through MATLAB MCP Core Server and MATLAB Agentic Toolkit.

Avinash Nehemiah, Head of Product Management and Marketing, Design Automation at MathWorks, outlined the company's view of the market.

"Engineering teams now have access to capabilities enabled by generative AI, and leaders need confidence that these translate into tangible engineering and business benefits," said Avinash Nehemiah, Head of Product Management and Marketing, Design Automation, MathWorks. "In engineering design and software verification, productivity improvements cannot come at the expense of rigour, traceability, or trust. MathWorks is committed to delivering grounded AI tools for engineering that help teams move faster while preserving the discipline and confidence required to develop complex engineered systems."

Broader updates

The update extends beyond AI features for embedded development. In education software, MathWorks has introduced MATLAB Course Designer to help educators build courses, labs, assessments and other teaching materials using MATLAB and Simulink.

For model exchange work, it has added Simulink FMU Builder, which creates standalone Functional Mockup Units from Simulink models and from C or C++ code to support integration workflows.

Several core MATLAB changes are also included. Users can build and share interactive webpages with visualisations without installing MATLAB, and engineering teams can manage Python environments and improve data exchange between MATLAB and Python workflows.

In Simulink, simplified context menus are intended to make common actions easier to reach. Users can also simulate C and C++ code within models without language limitations or additional wrappers.

Other product updates cover wireless communications, testing, mapping and signal processing. Wireless Network Toolbox now supports the modelling, simulation, analysis and visualisation of wireless communication networks to assess end-to-end system behaviour.

MATLAB Test has been updated so users can generate starter tests, equivalence tests and tests from command history using MATLAB Copilot. It also adds an option to run tests related to the current file, intended to reduce unnecessary execution.

Mapping Toolbox gains 3D building visualisation, image overlays and raster map functions, while Signal Processing Toolbox adds new Filter Designer and Filter Analyser apps, along with tools for labelling time-frequency data and extracting signal features.

Founded in 1984, MathWorks employs more than 6,500 people across 34 offices worldwide. MATLAB and Simulink are used in industries including automotive, aerospace, energy and medical devices, as well as in universities and research institutions.