Progress AI from DroneDeploy interprets real-world visual data from drone flights and 360 walkthroughs
Reality capture specialist DroneDeploy has launched Progress AI, a ‘visual intelligence’ solution for automated progress tracking from both aerial and ground capture data.
Progress AI uses vision-language models (VLMs) – a step beyond traditional large language models (LLMs) – to interpret real-world visual data from drone flights and 360 walkthroughs and then generate ‘structured insights’ automatically. According to the developers, this ‘next-generation’ AI doesn’t just read text; it understands what it sees.
“With Progress AI, superintendents can get a complete view of progress across every floor, every trade, in minutes, not hours,” said James Stripe, chief product officer at DroneDeploy. “Beta customers have already used Progress AI to track project milestones, validate pay apps and reduce trade conflicts without adding headcount or complexity. We expect these capabilities to scale rapidly as the solution rolls out across more project sites.”
In addition to automatically generating structured progress reports, Progress AI allows teams to ask natural-language questions about site conditions using voice or text. From superintendents verifying trade progress to project executives validating schedule risk, answers are delivered instantly – without waiting for status meetings or chasing reports.
“The superintendent can simply ask, ‘Where are we behind on drywall this week?’ and Progress AI will surface the facts directly from last night’s site capture,” added Stripe.
According to DroneDeploy, one of the key benefits of Progress AI is that BIM and schedule data are not required. Walk or fly your site as usual, the AI does the rest, the company claims.
Progress AI is powered by DroneDeploy‘s proprietary visual intelligence dataset which comprises billions of square feet of ground-level data and billions of acres of aerial imagery, delivering ‘high-resolution, survey-accurate visuals’ used across construction, energy, infrastructure and agriculture.
The AI model is trained to detect installed work, track changes and quantify progress with what DroneDeploy describes as ‘human-level accuracy’, continuously improving as more data is captured.
“Progress AI is like having an extra superintendent – capturing everything, seeing everything, analysing everything, so nothing gets missed,” said Cayman Wilson, Project Engineer at Wharton-Smith Construction Group. “I just walk the site like usual and Progress AI gives me a clear, visual heatmap of what’s happening and what’s changed.”
“We caught a framing issue with a door install just by reviewing the Progress AI walkthrough. That saved us hours – maybe days – of rework.”
Progress AI is available for early access now through October, when it will be formally available as an add-on for all DroneDeploy Aerial and Ground customers.
