Changing my own mind about fluoride by almost shitposting about a water filter startup

I was ready to absolutely tear into Water2.

After reading Entrepreneur.com's profile of the UK water filter start-up co-owned by Bear Grylls, one detail had me properly fuming: their Fluoride Filter Add-On. According to founder Charles Robinson, this product was born from Facebook comment sections where customers were "crying out for the product, almost begging to give us their money for it."

My immediate reaction was visceral (and probably a bit sanctimonious, tbh). How could a company sell something that removes a proven beneficial substance from drinking water? Fluoride prevents tooth decay, particularly benefiting children and lower-income communities. In my mind, the anti-fluoride crowd are typically the same folks worried about chemtrails and 5G towers.

I was prepared to write about how Water2 had failed to maintain a basic "Do No Harm" principle: companies should at minimum ensure their products don't actively hurt customers. I had my narrative locked and loaded, then I made the rather inconvenient mistake of actually diving deeper into the evidence.

The assumption that wasn’t

Like most people with a passing interest in public health, my position on fluoridation was straightforward: it's safe, effective, and the science is settled. The opponents are conspiracy theorists; the supporters follow evidence-based medicine. So when I saw Water2 selling fluoride removal, my brain filed it under obviously-harmful-snake-oil-being-peddled-to-gullible-wellness-types.

Before putting pen to paper, I wanted robust sources to back me up. I turned to our own work at Ambitious Impact (previously known as Charity Entrepreneurship), where I knew my colleagues did some research into whether it’d be worth to recommend salt fluoridation for tooth decay prevention as a non-profit start-up idea. These are some of the most trustworthy, thorough and smart researchers I know, so I was taken aback by their conclusion: "Our view is that scaling up salt fluoridation in LMICs is not an idea worth recommending to charity founders at this moment."

I had to know more.

The science

Here's what completely upended my smug assumptions:

Fluoride might be harmful at any level. Unlike vitamins that are beneficial at low doses but harmful at high ones, fluoride may follow the "heavy metals model", where any exposure carries some risk. AIM notes that fluoride is "not an essential mineral" and "our body does not need it to perform any physiological functions." It might be more like lead than like vitamin D.

The risk-benefit calculation is murky. AIM’s cost-effectiveness modelling suggested fluoride's potential negative effects could be "greater than the positive intended effect" on dental health.

The neurotoxicity evidence is genuinely concerning. Multiple systematic reviews found associations between fluoride exposure and reduced IQ in children. Some evidence suggests effects beginning at levels close to those used in water fluoridation. From the AIM report:

In our view of the evidence, there is a non-negligible chance that exposure, to even small amounts of fluoride, during pregnancy and childhood may lead to neurological damage, with a potential population-wide reduction in IQ of one-two points. While the evidence in this space is largely of low quality, and may be highly biased, we find this risk of doing harm quite concerning.

So much for settled science.

Some lessons from getting the wrong end of the stick

All this genuinely gave me pause to think. Part of me enjoys the moments when my thinking is actively being changed (the feeling of becoming better informed is always nice), but it’s also internally embarrassing to some degree. My final thoughts:

1. I will still drink fluoridated water

This uncertainty genuinely shook my confidence, but it didn't flip my position entirely. The concerning evidence mostly involves higher exposure levels, and the dental health benefits remain well-documented. So while I now think the science on negative effects is in flux, that’s not to say that the science on positive effects is also unsettled. While AIM’s report above does make the comparison to heavy metals, we’re not actually talking about heavy metals, so we don’t need a lead-in-turmeric-level freak-out. My money is still on the “fluoridation is likely net beneficial” horse, until I see some concrete evidence to the contrary.

2. Someone really needs to sort this out

The fluoride question shouldn't be this ambiguous in 2025. If there's genuine scientific uncertainty about a substance millions consume daily, that seems worth resolving definitively rather than leaving it to Facebook comment sections. The documented delays in publishing government research and the harassment of scientists studying fluoride neurotoxicity are particularly concerning.

3. My “Do No Harm” critique still stands (somewhat)

Yes, I misjudged the science. But, I still doubt Water2 arrived at their decision through careful evaluation. Their explanation (Facebook comments and customer demand) suggests they saw market opportunity rather than conducted thoughtful analysis.

Responsible business practice in scientific uncertainty should include acknowledging complexity rather than oversimplifying for marketing, investing in research rather than exploiting uncertainty, and transparently communicating evidence to customers.

It's the difference between "customers want this, so we'll build it" and "customers want this, so let's understand whether it actually helps them."

4. Market solutions can be a good hedge against uncertainty

Instead of blanket approaches like Utah's complete fluoridation ban, having market solutions like Water2's filter might be a sensible hedge against scientific uncertainty.

Yes, it's unequal—only people with disposable income can afford to opt out. But when experts genuinely disagree, giving people choice seems more reasonable than imposing one-size-fits-all policies. At least people can make their own reasonable risk assessments (e.g. if you’re pregnant, it’d be very reasonable if you wanted to buy Water2’s filter for the duration of your pregnancy).

A different conclusion from the one I expected

Perhaps the most valuable lesson is about intellectual humility. I started this research absolutely convinced I was right, armed with what I thought was settled science.

The evidence forced me to conclude I didn't know what I thought I knew.

This happens more often than we'd like to admit. Since I began my career in global health and development I’ve encountered this repeatedly, changing my mind a number of times about different interventions that seem obviously beneficial until you dig deeper.

The solution isn't paralysing relativism where all positions are equally valid. It's developing better self-questioning habits, reading deeper and more widely, and maintaining enough humility to change our minds when evidence demands it. Often, the most responsible thing is admitting you're an idiot.