A Letter to the Editor
It was great to read the last edition of e-chronicle and hear all the different views on the application of AI, as well as its risks and opportunities. I’ve been discussing this a fair bit with colleagues, friends and family – so I thought I’d put pen to paper (or in my case – single digits to keyboard) and offer my two cents.
As AI rapidly finds its way into our industry (and everyday lives), whether we want it or not, I think it’s important we still consider one of the central tenets of the environmental industry—the Precautionary Principle—and why it is critical to consider the uncertainty involved with AI.
It seems easy to simply accept and adopt this new technology and a new style of working. In fact, it is almost harder NOT to accept it. But we shouldn’t ever forget that our science is built on curiosity – and our industry and practice is based on ‘first-principles’ thinking.
I like to think of first principles simply as the ability to take something apart and put it back together, based on observations and foundational knowledge. In our industry, and as my colleague Phil Mulvey likes to say – it’s understanding the presence of contamination, understanding the environment and then predicting the interaction between the two.
Humans have changed our environment substantially over the past 8,000 years. However, since the Great Acceleration of the mid-20th century, we increasingly live and work in a “plastic environment”— surrounded by artificial and synthetic materials. (Radiohead were certainly on the money with their song Fake Plastic Tree’s in 1995!).
It seems to me that we’re now also starting to surround ourselves with synthetic information – or ‘AI Slop’ as it’s helpfully been badged. Data can be copied, summarised, inferred and re-packaged so effortlessly that it risks becoming “plastic data”, flexible enough to support almost any narrative.
AI is remarkably fluent (and sycophantic), but fluency can also be mistaken for accuracy.
The contaminated land industry has seen this before. Chemical suppliers once promised “Better living through chemistry”, but so did Fatboy Slim (there’s a theme here). They offered ‘wonder’ products that promised convenience, comfort and a better life, solving immediate problems but creating environmental problems that emerged decades later.
AI presents a similar challenge. It promises convenience, speed and productivity, but with hidden costs The environmental footprint of data centres is not yet fully apparent – and as the exponential volume of data generation grows, so too the demand for energy and water. Another cost that we may not have fully considered could be the gradual de-skilling of the very professionals our industry expects to exercise judgement.
Perhaps one of the greatest risks is not that AI will make mistakes, but that experienced practitioners stop recognising them. When AI confidently suggests the next prompt or the next recommendation, it subtly shifts from being a tool to becoming a guide. The danger is that our thinking becomes reactive rather than investigative, and we lose our sense of inquiry.
Contaminated land assessment has always involved ambiguity. Two competent practitioners can review the same dataset and legitimately arrive at different conceptual site models because interpretation requires context, experience and professional judgement. No algorithm can replace healthy skepticism or a practitioner willing to ask, “What if this assumption is wrong?”
Ironically, we have invested enormous effort in training AI models to recognise patterns (reCAPTCHA anyone?), yet risk spending too little time deliberately training people to do the same. Pattern recognition, critical thinking and curiosity are skills developed through experience, mentoring and questioning—not merely through exposure to technology.
Perhaps the future of our profession is not choosing between humans and AI but ensuring that AI never replaces the habits that made good, contaminated land consultants in the first place. We should embrace tools that enhance our capability while resisting those that erode our curiosity.
In contaminated land consulting, first principles are not old-fashioned; they are our quality assurance system. And curiosity is still the most valuable instrument in every consultant’s toolbox.
Ok – I’ll hop off my soapbox now…