Artificial Intelligence is it here to make our lives easier, or take our jobs away? Is it the next step in technological advancement, or a fad? And when will it steal Arnold Schwarzenegger’s face and demand we all serve the whims of Skynet?
The reality is that AI is everywhere – and has been for a while.
From the phone in your pocket to the golf clubs in your trunk, AI technology is being integrated into every facet of our daily lives to increase efficiency and eliminate mistakes caused by human error.
It’s no different in the construction industry, where AI is assisting in project planning, Operations and Maintenance (O&M), jobsite safety, and more.
Eric Sullivan is the Director of Business Development at SewerAI, a company founded in 2019 with the mission to leverage AI and cloud services to, as he puts it, “enable our customers to achieve superhuman results.”
SewerAI sells software that assists in the inspection of wastewater infrastructure – a vital sector of the construction services industry, given that the United States has over 6 billion feet of underground sewage pipe and, according to the United States Environmental Protection Agency (EPA), half of that infrastructure is at the end of its useful life and failing.
Routine inspections of these systems using remote-controlled, CCTV camera-equipped rovers and push-fed sewer scopes can identify issues that, if otherwise left undetected, would cause service interruptions and other headaches for customers and utility owners alike.
Currently, the industry standard is for this process to be conducted by a technician who observes the rover’s footage in real time while on site, noting every defect as they navigate the device through the sewer system. The average CCTV sewer inspection truck features a small computer lab in the back to accommodate this on-site evaluation, which usually takes place using a software such as WinCan – another company that’s now offering AI-powered inspection solutions.
Sullivan and his team believe that introducing AI tools into the inspection process increases the efficiency and accuracy of these investigations. SewerAI’s software notes defects automatically as the rover traverses the waste or stormwater system, eliminating the need for a technician to stop and start the inspection to note these findings.
“[We’re] trying to provide the best total solution for people as it relates to inspection data capture,” he said. “Getting it to a place where people can collaborate easily around it, but also getting into an environment where different AI tools can be applied to speed up the process of cities and engineers getting condition assessment deliverables, make that process cost less and also, in the end, achieve a more consistent, better deliverable, more accurate coding, and more accurate assessment.”
AI Fact vs. Fiction
Artificial intelligence technology has a long way to go before it’s ready to hunt Sarah Connor across Los Angeles.
Unlike the near-future dystopia depicted in the Terminator film franchise, today’s AI-driven tools – from Apple’s Siri to OpenAI’s ChatGPT – are built on traditional machine learning models that still rely on algorithms developed and maintained by data scientists.
The humans still control the robots – at least, for now.
“AI capabilities have been evolving steadily since the breakthrough development of artificial neural networks in 2012, which allow machines to engage in reinforcement learning and simulate how the human brain processes information,” IBM’s Data and AI Team wrote in an article from October 2023. “Unlike basic machine learning models, deep learning models allow AI applications to learn how to perform new tasks that need human intelligence, engage in new behaviors and make decisions without human intervention. As a result, deep learning has enabled task automation, content generation, predictive maintenance and other capabilities across industries.”
Sullivan said that SewerAI’s software is largely based on a technology known as computer vision, which combines cameras, edge (local device) and cloud-based (remote server) computing, software, and machine-learning AI to help systems “see” and identify objects.
“We’re solving one very specific, but important problem… [Our products] enable a workflow change that has been largely embraced in the industry over these last few years,” Sullivan said. “Now, an operator can simply put a robotic camera crawler into a pipe. They can make sure that the recording is on, the lights are good, that the camera is centered in the pipe, and they can collect a lot more data in a day inside of that workflow compared to traditional, manual data entry.”
Similarly, Autodesk, Inc.’s Autodesk AI allows users in Design and Make industries to automate 2D documentation, automate certain tedious design tasks, perform predictive analysis for wind, noise, and operational energy in real time, and more.
“As the trusted technology partner for Design and Make industries, Autodesk sees AI as a way for our customers to tackle the challenges they face and turn them into opportunities,” Autodesk President and CEO, Andrew Anagnost, said in a press release announcing Autodesk AI. “Autodesk AI is the assistant that lowers the learning curve for designers and engineers, supercharges their productivity, and accelerates innovation.”
Experts believe that AI can even help improve safety on job sites – something desperately needed in the construction industry – which in 2022 saw the third-highest fatality rate among relevant sectors, with 9.6 deaths per 100,000 full-time workers.
Boston, Massachusetts-based construction company Suffolk was recently featured in a Business Insider series exploring how AI is being used across different industries. In 2017, Suffolk appointed its first chief data officer, Jit Kee Chin, and invested in construction startups and used its own jobsites to test out potential solutions to safety-related problems.
Suffolk worked with NewMetrix, a startup based in Cambridge, Massachusetts, to develop a safety-solution-leveraging AI to digitize the process of safety observations and incident data gathering.
“The shift from manual and analogue to digital enabled us to collect data, and that’s the most important piece of all of this,” Suffolk’s National Director of Operational Excellence, Kelsey Gauger, told Business Insider. “It plays into our data strategy.”
NewMetrix then developed a predictive model that leverages AI to assign risk ratings to Suffolk job sites and predict the likelihood of an incident occurring on them. The model’s findings led Suffolk to change their key performance indicator (KPI) for jobsite safety: from a target of reducing the total number of on-site incidents to the number of safety observations carried out per workforce hours accumulated.
Suffolk saw a 25% improvement rate in their total recordable incident rates in the fiscal year following the implementation of this data-driven strategy. NewMetrix’s machine-learning model now accounts for 40 different factors correlated with safety outcomes on Suffolk job sites.
“One of the things that we’ve been able to do is actually flag at-risk projects, in addition to driving safety observations, so we can actually plug into that model all of this information and it can tell us ‘Here are the three jobs in your portfolio that are the most at risk,’” Gauger said.
Flipping the Question of Job Security
Sullivan acknowledges that SewerAI’s name is as much a blessing as it is a curse in a world captivated by the possibilities of artificial intelligence. It places them on the cutting edge of construction technology – but it also conjures visions of a world that, in truth, is still strictly science fiction.
“We’re using technology that’s been around for about 15 years, honestly,” Sullivan said. “All of us have uploaded photos to Facebook at some point in time, and [Facebook] was like ‘Hey, that’s you, Eric, and your friends.’ This technology has been around a long time, and it took a minute for our industry [to catch up.]”
AI isn’t going to take over the world tomorrow, but many employees are concerned that it’s going to take their jobs away.
Resume Builder recently surveyed 750 business leaders at companies that currently use or plan to use AI in 2024:
- 53% of the companies surveyed use AI, and 24% plan to start in 2024
- 37% of the companies using AI say the technology replaced workers in 2023
- 44% of companies surveyed say AI will lead to layoffs in 2024
- 96% of companies hiring in 2024 say candidates will benefit from having AI skills
- 83% say AI skill will help current employees retain their jobs
AI can’t steal jobs that were already open.
There were 413,000 open construction jobs on the last day of January 2024. That’s more unfilled positions than the same time in 2023, and near the record 454,000 vacancies that existed the previous November, according to data from the Bureau of Labor Statistics.
Henning Roedel, robotics lead for the innovation team at Redwood City, California-based DPR Construction, told Construction Dive in an article published last May that, “We don’t think about how to reduce our staff size, because we have enough backlog and work ahead of us that we need more people.”
“You need to flip the displacement question around because we currently don’t have enough people in our industry to meet the construction needs of society as it is,” Roedel continued.
Sullivan said that he has yet to see SewerAI’s technology cost someone their job. He does acknowledge that an experienced sewer inspection technician might see their position eliminated as computer vision eliminates the need for manual, real-time coding of sewer defects. However, he argues that those adept at this type of work generally are promoted out of it rapidly anyway as their skillset is useful higher up in a sewer inspection company’s hierarchy.
“We actually hire a lot of people that are wonderful CCTV truck operators to do what they do here, which is use the AI tools,” Sullivan said. “We still have people on our team that need to review the AI output, watch the source data video playback themselves, make some very important interpretations and distinctions that AI is not able to make yet… What we do is we enable an AI-assisted process…”
While it’s not likely to replace a workforce anytime soon, AI does face other challenges as its champions look to integrate it into the construction industry. Many of these concerns aren’t industry-specific; questions about data privacy and technology cost arise no matter which sector is looking to introduce AI-driven solutions to its problems.
“Despite these challenges, the potential of AI in construction is immense,” American Management Services, Inc.’s Louis Mosca wrote last year in an article in Forbes Magazine. “As technology continues to evolve, we can only imagine how much more it could reshape the industry.”
Producers of AI-driven tools for the construction industry believe they are doing what has always been done: innovate to increase productivity, efficiency, and accuracy.
“How can you get your workforce to perform better, have an easier time, maybe even make the job site safer, and do it quicker and with more consistency with less risk on your end?” Sullivan said. “These are all the things that people are interested in, generally. So that’s the type of feedback that we interact with is like ‘Oh, ok, there’s a business case here for this.’”
