Bentley launches new AI tools & collaborative initiative to transform engineering

Bentley Systems has introduced new artificial intelligence (AI) capabilities designed to support infrastructure engineering, along with a new collaborative initiative involving industry partners.

AI adoption and survey findings

Bentley highlighted that AI use is increasing across infrastructure engineering, citing its Going Digital Awards, where nearly one-third of submissions and nearly half of finalists used AI within their projects.

These findings match the results of a global survey of infrastructure professionals, conducted by Bentley with collaboration from law firm Pinsent Masons, engineering firm Mott MacDonald, and consultancy Turner & Townsend. The survey found about half of respondents are either piloting or have implemented AI, with further plans to expand its use. Respondents identified productivity in design and engineering, and automating documentation, as key areas of focus.

"AI is poised to transform infrastructure," said Bentley CEO Nicholas Cumins. "At Bentley, our vision is for AI to empower infrastructure engineers-not replace them. Trustworthy AI, built on infrastructure context, can improve engineering productivity and transform workflows across project and asset lifecycles."

Cumins further noted, "The greatest challenge to delivering better and more resilient infrastructure is engineering capacity. The reality is, there simply aren't enough engineers in the world to do all the work that needs to be done. AI promises a step change in productivity that can help close this capacity gap."

AI in Bentley's products

Bentley has expanded AI features across its product range, following the previous launch of OpenSite+ for civil site design. The company has now launched products for substation design and construction management, intended to accelerate AI adoption in project delivery.

"We have been creating a new generation of infrastructure applications built on digital twins, powered by AI, and fully connected to Bentley Infrastructure Cloud," said Francois Valois, Senior Vice President, Bentley Open Applications. "They show what's possible when AI is built for real infrastructure workflows and tailored to the needs of engineers."

Ongoing AI integrations include:

• OpenSite+ - A generative AI tool for civil site design, available in limited release, claimed to deliver projects up to 10 times faster while maintaining accuracy.
• OpenUtilities Substation+ - An AI-powered digital twin application for substation design, enabling multi-user collaboration in a single model. An early access programme will begin in November.
• SYNCHRO+ - An AI-powered construction management platform, offering 4D modelling, integration with geospatial provider Cesium, and automated construction sequencing. Early access will start in December 2025.

Bentley plans to introduce AI agents that will automate annotation in OpenRoads Designer and OpenRail Designer, set for release in November 2025, with Bentley Copilot integration in early 2026.

ProjectWise, part of Bentley Infrastructure Cloud, will see new AI-powered search capabilities aimed at reducing time spent finding information, with early access in December 2025 and full release planned in 2026.

Brianne Belschner, Model-Based Design Lead at VHB, who worked with Bentley on OpenSite+, commented, "OpenSite+ is going to open a lot of doors for us to move faster, get better answers, and really be better engineers."

Data stewardship and governance

Bentley reaffirmed its commitment to data stewardship and intellectual property protection amidst these AI developments.

"Our users are in control of their data. They decide if it is used for AI training, and to what extent," Cumins said. "Our users' data is their data, always."

Bentley uses data for AI training only with explicit permission or purchase, and offers customers the ability to fine-tune AI models with their own data for exclusive organisational use. The company has introduced a Data Agreement Registry to give users transparency around how data is used for AI model training.

Trustworthy and contextual AI

Bentley emphasised that AI in infrastructure engineering must be grounded in real-world context, and tailored for a sector where precision is critical.

"Infrastructure engineers work in a creative profession, but one where precision is non-negotiable and consequences are real," Cumins explained. "That's why AI in infrastructure must be grounded in real-world context."

Bentley noted that companies are using Bentley's tools to inform their own AI models, combining design data, subsurface information, and engineering principles from the company's platforms.

Cumins added, "What's striking is how diverse these AI use cases are. AI is emerging across design, construction, and operations, touching every phase of the infrastructure lifecycle. And the results go well beyond time savings; they're using AI not just to automate, but to optimize decisions and outcomes in ways that were not possible before."

The company cited a geothermal project in Turkey that reduced its timeline from five years to one by leveraging AI and GPU-accelerated simulation, carrying out over 10 million scenario evaluations in days and 3,000 simulations in hours, achieving costs reductions of over 75%.

AI co-innovation initiative

Bentley announced an "Infrastructure AI Co-Innovation Initiative" to collaborate with engineering firms and asset owners in developing the next generation of AI workflows, and will consider how APIs and commercial models can evolve for AI and human-driven work.

"For more than 40 years, we've helped infrastructure professionals and organizations become more productive through our software," Cumins said. "We welcome the creative ways our users are already combining our applications with AI-and we believe this is just the beginning."

"This is a pivotal moment," Cumins said. "The opportunity to shape the future of infrastructure is in front of us-and we're incredibly excited to collaborate with our users in this new way."