Dr. Maor Farid, Co-Founder & CEO at Leo AI
Jan 1, 2026
AI in mechanical engineering means different things depending on who you ask. Is it a chatbot? Will it design entire cars by itself?
The landscape of industrial design is shifting under our feet. For decades, mechanical engineering relied on a linear progression of tools—from drafting boards to 2D CAD, and eventually to the sophisticated 3D modeling and PLM systems we use today. But we are now standing at the precipice of a new era: the age of AI in Mechanical Engineering.
In our latest webinar held just a few weeks ago, we sat down with Ayora Berry, the VP of AI Product Management at PTC, to cut through the noise surrounding Artificial Intelligence. Is it just a chatbot? Will it design entire cars by itself?
Ayora mapped out the trajectory of AI for Engineering not as a sudden explosion, but as three distinct time-based horizons. If you missed the live session, this summary covers the critical insights you need to understand to stay competitive in 2026 and beyond.
Here's what engineering leaders need to know.
Horizon 1: Turbocharging the Worker (The Advisor)
Ayora explained that we are currently living in the first horizon. In this stage, AI functions primarily as a "Turbocharger" for the individual engineer. It does not replace the human; it acts as a capable advisor.
If you have ever spent hours digging through a PDM system looking for a specific bolt used in a legacy assembly three years ago, you understand the pain point here. During the webinar, it was noted that engineers can spend nearly 6 to 8 hours a week just searching for information. That is an entire workday lost.
AI CAD tools in this horizon solve this by understanding context. As Ayora Berry noted, "Context is king." Unlike a generic large language model (LLM) that might hallucinate a part number, an engineering-focused AI understands the "boundary representation" (B-rep) of your 3D models. It knows that if you are designing a suspension system, you need parts that fit specific load and geometric constraints.
Horizon 2: Enterprise Intelligence (The Assistant)
The second horizon, expected to mature over the next 1 to 3 years, moves beyond the individual. This is where we see the rise of AI for engineering at the enterprise level. The goal here is to elevate the "Enterprise IQ."
Mechanical engineering has long suffered from the "silo problem." The design team uses CAD; the electrical team uses ECAD; the software team uses ALM; and manufacturing uses ERP. These systems rarely talk to each other fluently.
Ayora described a future where AI acts as an assistant and an orchestrator. Imagine a scenario where a design change is proposed. In the traditional workflow, this triggers a manual chain of emails and meetings.
With Agentic AI (a concept PTC is exploring in partnership with Microsoft) AI agents can "talk" to one another across these silos. A "Change Impact Agent" could autonomously fetch information from the ERP regarding inventory, check the PLM for legacy lessons learned, and present a consolidated report to the human engineer.
Horizon 3: Reinventing Product Development (The Automator)
The third horizon (3 to 7 years out) is the most transformative. This is the stage where AI in mechanical engineering begins to change how products get built.
Currently, we follow a V-model of systems engineering: define requirements, design, build, and then validate. Validation often happens late in the game, where mistakes are expensive. Horizon 3 envisions a world where AI helps "shift left."
In this future, AI isn't just an advisor; it is a co-pilot capable of generative tasks. An AI system might look at a requirement for a new gearbox, analyze historical versioning, and automatically generate a CAD geometry that meets those requirements. It could then run its own simulations to validate the design before a human even rotates the model on the screen.
The Friction: Competition vs. Expectations
During the Q&A portion of the webinar, we discussed the friction in adopting these AI CAD technologies. Interestingly, Ayora pointed out that the motivation differs depending on where you sit in the organization:
For Engineering Leaders (VPs, CTOs): The driver is competition. They know that if their competitor adopts AI-driven generative design, they will be able to launch products 30% faster.
For Individual Engineers: The driver is expectations. The standard for "productivity" is rising. The fear isn't necessarily losing a job to AI, but losing a job to an engineer who uses AI effectively.
How to Prepare: Data Hygiene
Ayora left the audience with pragmatic advice: Get your data house in order.
AI is only as good as the data it feeds on. If your PLM is a mess of unstructured folders and duplicate files, the AI will struggle to provide accurate context. Smart organizations are starting with low-risk, high-value "Advisor" use cases today to build trust, while preparing their data infrastructure for the high-risk "Automator" use cases of tomorrow.
This post only scratches the surface of the conversation we had about the future of industrial software. If you want to hear the full discussion with Ayora Berry and see exactly how PTC is approaching these challenges, watch the full webinar interview below.
[Watch the Full Webinar Interview Here]
The challenges Ayora described in Horizon 1, particularly the hours engineers spend searching for information, are exactly what we built Leo to address. If you're curious how that works in practice, reach out to our team.
