AI is getting better quickly. But for those in the Design & Make industries, a capable model is only part of the equation.
To be genuinely useful, AI needs to understand the work behind a request: the project, the data and tools involved, the decisions already made, and how they connect. It needs to work with the specialized technology and expertise organizations already rely on. And it needs to draw on different AI models for different tasks.
At Autodesk University 2026, we shared how we’re addressing those challenges as we continue to evolve Autodesk Assistant.
Autodesk Assistant starts with what you’re trying to achieve and helps you move the work forward. Rather than requiring you to navigate products, workflows, and commands on your own, Assistant brings forward the right Autodesk capabilities to help turn your intent into work you can review, refine, and use.
This year, we’re expanding what makes that possible: giving Assistant more context about the work, connecting it to a broader range of AI and specialized capabilities, and bringing the right AI to each task.
Autodesk context helps AI understand the work and the decisions behind it.
Autodesk Assistant Builder will let organizations connect their own and partner AI, tools, and expertise to Assistant, while Autodesk AI Orchestrator will select the right model for the task, so you don’t have to manage models behind it.
Together, they help make Autodesk Assistant more powerful while keeping professionals in control.
Grounding AI in your work
Your work depends on context.
Knowing that a wall moved, a component changed, or a schedule slipped is useful. Knowing why it changed, what else that decision affects, and what was happening when it was made is far more valuable.
Autodesk context helps provide that foundation.
It brings together relationships between project information, people, decisions, and activities. This includes not only what happened, but important context around decisions that were made and when they occurred within the workflow.
For Autodesk Assistant, that means responses can be grounded in relevant project context, not just the information someone includes in a prompt. Professionals can spend less time searching for what happened, retracing decisions, and piecing together how changes affect the work.
Autodesk Assistant Builder: Connecting Assistant to your organization’s intelligence
The knowledge that powers your work doesn’t live only in Autodesk tools.
Organizations rely on their own data, systems, AI, and expertise, along with specialized technology from partners. Assistant becomes more useful when it can work with that broader ecosystem.
Autodesk Assistant Builder brings that broader ecosystem into Autodesk Assistant.
With Assistant Builder, organizations can connect their own AI and partner AI alongside Autodesk capabilities, so Assistant can draw on the tools, data, and expertise they already use.
The goal is straightforward:
Your AI. Your partners’ AI. Autodesk AI. Working together through Autodesk Assistant.
This means Assistant can work with more of the systems, knowledge, and expertise your teams already rely on, rather than requiring everything to live within Autodesk tools.
For developers and partners, it creates new ways to extend Assistant with specialized tools, data, agents, and capabilities.
Autodesk AI Orchestrator: Bringing the right model to the workflow
There is another layer of complexity customers shouldn’t have to manage: the AI models themselves.
No single AI model is best at every Design & Make task. Working with geometry is different from analyzing documents. Simulation, engineering, images, and schedules can each require different AI capabilities.
Meanwhile, the model landscape continues to change rapidly.
Autodesk AI Orchestrator is designed to manage that complexity behind the scenes.
Autodesk analyzes specialized Autodesk models and frontier models for factors including accuracy, speed, and cost. Orchestrator then selects the right model, or combination of models, for the task.
Customers don’t need to determine which model should handle every step or keep up as models change. They start with what they want to accomplish, while Autodesk AI Orchestrator manages that complexity behind the scenes.
How it comes together
These capabilities are designed to work together, because useful AI needs more than a model. It needs context about the work, access to the right capabilities, and a way to choose the right AI for the task.
You start with what you want to accomplish, and Autodesk context grounds Assistant in the work. Assistant Builder expands what it can draw on. Autodesk AI Orchestrator selects the right model for the task.
Together, they move Autodesk Assistant toward an AI-first experience built around what you want to accomplish, not the products, commands, or models you need to get there.
Today, Autodesk Assistant is embedded in Autodesk products, helping customers answer questions, get guidance, and complete product-specific tasks. At AU, we previewed the next step: a standalone Assistant experience for work that crosses products, projects, and teams.
These are connected experiences, not replacements for one another. Over time, they will increasingly share context, capabilities, and intelligence, giving customers different ways to work with Assistant depending on what they’re trying to accomplish.
Across both experiences, professionals remain in control. Assistant can produce information, recommendations, and work, while professionals inspect, refine, approve, and decide what gets used.
Autodesk Assistant is evolving around a simple idea: start with what you want to accomplish, then bring the right context, capabilities, and AI together to help you get there.