Trimble’s Stuart Campbell, explores AI in construction, the call for standardisation and where the responsibility should lie
AI has been part of our technological landscape for years, but its capabilities are rapidly advancing. The construction industry, in particular, is witnessing unprecedented changes driven by AI, with technology being used in ways unimaginable just a year ago. As AI continues to evolve, it promises to drive significant improvements in industry operations.
Generative AI stands as a prime example of this evolution. The advent of tools like ChatGPT has revolutionised our interaction with AI, integrating it into daily tasks such as automated meeting transcriptions and suggested email responses. These tools help to enhance productivity by allowing us to instead focus on tasks that require expertise and judgement. Generative AI is also paving the way for advancements in generative design and machine learning.
In essence, AI enables machines to handle repetitive tasks, freeing humans to leverage their specialised skills. AI is not a replacement for people but a complement that enhances human capabilities.
Benefits vs challenges
Whether you view AI optimistically or pessimistically, its benefits in boosting productivity and efficiency are undeniable. AI minimises mundane tasks, allowing professionals to focus on higher-level work that requires human insight and experience. According to the McKinsey report “Artificial intelligence: Construction technology’s next frontier”, if widely adopted, AI could significantly improve the construction industry’s productivity levels.
Either way, the last year has raised AI up the agenda for the built environment, making people curious and willing to engage with it.
However, the rapid development of AI also brings challenges. The need for responsibility and risk management in AI adoption is paramount. As the industry embraces AI, there is a pressing need for formal standards and guidance on AI implementation. Questions concerning where and how AI should be applied, and how its use should be communicated to clients, remain critical.
Industry feedback indicates a clear need for balance between AI’s risks and opportunities. Striking this balance requires an understanding of AI’s capabilities and limitations and the potential risks it presents. Organisations must be prepared with AI strategies aligned with their digital and data initiatives.
The call for standardisation
The construction industry is calling for formal standards and guidance to build confidence in AI adoption. This need is evident in sectors like the nuclear industry, where accuracy, reporting, and traceability are crucial. The potential security threats and risks of error make it challenging for new technologies to penetrate these sectors. However, when correctly implemented, opportunities abound.
Responsibility and liability
A critical question in AI adoption is where the responsibility lies if something goes wrong. Whether you are a contractor, engineer, surveyor or consultant, you have to stand by the advice, service and output you provide to clients, and like all things, AI isn’t infallible.
Speaking to those in the industry and the consensus suggests that a company’s liability for AI is akin to its liability for any other product or service. This makes it the company’s responsibility to ensure quality checks before sharing AI-generated outputs. AI does not eliminate the need for human oversight; rather, it heightens the need for quality assurance.
For us at Trimble, quality data is essential from a technological provider’s perspective. The principle of “garbage in, garbage out” underscores the importance of feeding accurate information into AI systems for effective outcomes.
Ultimately, AI is a tool, and we remain the professionals who wield it. While AI can provide predictive and generative insights, human judgement is crucial in interpreting and applying these insights. Responsibility cannot be abdicated to AI; it remains with us, the users.
The future of AI in construction holds immense potential, but its successful integration depends on responsible use, guided by robust standards and informed by human expertise.
Stuart Campbell is senior manager, enterprise sales team at Trimble
