The Truth Foundation

Image: Paulasleepwalker via Flickr

Here are some things to know about truth: Truth exists. Finding truth takes work. And truth changes over time. I devote my energy to finding and promoting that truth, even when it is uncomfortable. It’s time for our biggest technology companies to do the same.

Myths are persistent

Megan Scudellari published an article in Science one year ago, called “The science myths that will not die.” Many of us cherish the beliefs that this article calls into question. For example, there’s this:

Screening saves lives for all types of cancer

Regular screening might be beneficial for some groups at risk of certain cancers, such as lung, cervical and colon, but this isn’t the case for all tests. Still, some patients and clinicians defend the ineffective ones fiercely.

. . . . “We’ve all been taught, since we were at our mother’s knee, the way to deal with cancer is to find it early and cut it out,” says Otis Brawley, chief medical officer for the American Cancer Society.

But evidence from large randomized trials for cancers such as thyroid, prostate and breast has shown that early screening is not the lifesaver it is often advertised as. For example, a Cochrane review of five randomized controlled clinical trials totalling 341,342 participants found that screening did not significantly decrease deaths due to prostate cancer1.

“People seem to imagine the mere fact that you found a cancer so-called early must be a benefit. But that isn’t so at all,” says Anthony Miller at the University of Toronto in Canada.

But early screening saved my life, I hear the breast cancer survivor saying. That is one truth for one person. But the medical truth is more complex, because early screening has negative effects as well, including unnecessary surgeries that bring their own risks.

And then there’s this cherished myth:

The human population is growing exponentially (and we’re doomed)

. . . the human population has not and is not growing exponentially and is unlikely to do so, says Joel Cohen, a populations researcher at the Rockefeller University in New York City. The world’s population is now growing at just half the rate it was before 1965. Today there are an estimated 7.3 billion people, and that is projected to reach 9.7 billion by 2050. Yet beliefs that the rate of population growth will lead to some doomsday scenario have been continually perpetuated. . . .

The world’s population also has enough to eat. According to the Food and Agriculture Organization of the United Nations, the rate of global food production outstrips the growth of the population. People grow enough calories in cereals alone to feed between 10 billion and 12 billion people. Yet hunger and malnutrition persist worldwide. This is because about 55% of the food grown is divided between feeding cattle, making fuel and other materials or going to waste, says Cohen. And what remains is not evenly distributed — the rich have plenty, the poor have little. . . .

“Overpopulation is really not overpopulation. It’s a question about poverty,” says Nicholas Eberstadt, a demographer at the American Enterprise Institute, a conservative think tank based in Washington DC. Yet instead of examining why poverty exists and how to sustainably support a growing population, he says, social scientists and biologists talk past each other, debating definitions and causes of overpopulation.

Truth is a slippery thing

When you have committed yourself to an idea, like early cancer screening or exponential population growth, it is very hard to give it up.

The same applies to widely discredited science myths that still have adherents, like vaccines causing autism.

Here’s the problem: truth is not fixed. Even in science, evidence changes over time. Our understanding improves, and we sometimes have to give up on ideas that used to make sense but on further scrutiny do not.

The question of truth is central in the world we now live in, where Facebook offers us biased, false news on a regular basis. It is too easy to believe what you want to believe, and too much work to examine the evidence. Facebook feeds encourage lazy thinking.


Science is imperfect, political, and often wrong, but it’s self-correcting and overall, moves knowledge forward. We would have no GPS, no smartphones, no pharmaceuticals, and no plastics without it. I believe in science, even if some of my scientific beliefs have had to change over time. (The idea that bacteria caused ulcers used to be absurd until it became the incontrovertible truth.)

Science and the skeptical thinking that accompanies it are the only cure for the plague of fake news — or call it what it really is, propaganda — that is sweeping over us. (While you can fight falsehoods from the other side with your own falsehoods, then we’ll all just be wrestling our way to oblivion.)

So here are some principles from science that can serve us as we read things:

  • Treat all facts as provisional. Ideals are constant (such as compassion for others, or a dedication to truth). But facts are subject to revision based on new evidence.
  • Seek out and challenge facts. If you see evidence that you may be wrong, run towards it, not away from it.
  • Prize learning over certainty. If you never change your beliefs, you are dead. If your beliefs are true, then new evidence can’t threaten them. If they are false, then why are you clinging to them?
  • Consider the source. It matters who says things. If it is The New York Times, The Bureau of Labor Statistics, or the Centers for Disease Control, it’s more likely to be right than wrong. If it is Breitbart or Occupy Democrats or The Food Babe, then you can’t trust it. Democrats and Republicans are both capable of bias — a balanced diet of different sources is the cure.
  • Don’t spread what you can’t verify. If you share it on Facebook just because you want it to be true, you’re a vector for the disease of falsehood. It doesn’t matter how much you want it to be true.
  • Learn the truth about statistics. Here are some things that should make you skeptical: small samples (a few dozen people), biased samples, statements about growth rates, and the amplification of small effects. Robust evidence comes from large samples that generate a consistent picture over time. Never trust a statistic without source. Those kinds of stats are wrong 93% of the time. (Just checking if you’re paying attention.)

The Truth Foundation

This stuff is work. It’s work that I’m willing to do, but many of us aren’t. Fake news and myths would not spread if we all put in the work, but most people lack the tools and the energy to think like skeptics.

It’s easy to blame our educational system, but that’s a cop out.

Our real educational system now is the Internet. So the Internet — or the companies that matter there — have a responsibility to help.

I call on Google, Facebook, Twitter, Microsoft, Amazon, Oracle, IBM, Cisco, Alibaba, Tesla/SpaceX, and Comcast to start an foundation for truth. Their resources are more than enough to educate people. Start a Truth Foundation that trains people in how to think clearly and understand. Dedicate your unused advertising units to publicizing it. Make it fun. Make it effective. Make it scientific.

If you need help, call me. I’ll put in as much time as you’d like.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.


  1. One thing I tell my kids ad nauseum: “Doubt and Uncertainty are our friends — they keep us on our toes; they remind us to be careful. It is when we are most certain that we are right that we are most likely to fall prey to error. Most of the great evil in this world is committed by people who are *absolutely sure* they are correct.”

  2. Many years ago, Lester C King, then a well-known professor of geology, told a class of budding geologists that “truth” is a matter for philosophers and theologians. For the scientist, there is only doubt.

  3. Josh, I think you have an editing error at the end of the “Robust evidence” sentence. Sorry to nitpick. I know you put these out on a deadline but removing even small mistakes leaves less room for others to challenge you.


  4. Philosophically, I have to take a small step sideways. Objective reality is definitely out there, but we have no way to get it unfiltered. Our perceptual apparatus, our language, and all of our memories and preconceptions get in the way of that. While our perceptions are merely ~= reality, they are based on some objective reality. We can get deep in the weeds about math, physics, etc. and how they ground us to reality, but human context is still a part of those discussions.

    Opinion is a different thing. It is not necessarily ~= reality. It may have no basis in fact or even speculation by people who study the subject deeply. Treating it as equivalent to reality or truth is, as you say, dangerous.

    As you point out, truth can be localized/specialized or universal. It can also be based on belief rather than evidence. I’m not sure if there are any languages with different words for internal vs. external truths like that. English (or American) just doesn’t do a good job in distinguishing the finer points most of the time.

    I hope truth, Truth, and reality prevail in the long term. I’m with you in working toward that goal.

  5. I love the idea, though the challenge for such a foundation would be not letting it get politicized. Every organization designed to collect and disseminate information gets politicized, most of them very rapidly. Fact-checking news organizations are the most embarrassing example of this, they continually pass conflicting judgments on essentially the same lie where preference is shown to their preferred political allies. This is not even nefarious, it is a natural side effect of our mental apparatus, which has to quickly sift things and uses all sorts of shortcuts to do so. This includes very smart and educated people, who in my experience can be the worst offenders because they have a tendency to think that their thoughts are automatically better by mere virtue of their education.
    A truth foundation sponsored by these big companies would also have to kowtow to their parents’ interests, whether commanded to do so explicitly or implicitly (in an IRS/Lois Lerner kind of way). That happens at my employer, it happens at every one of my clients. There’s simply no way around this: Nobody will bite the hand that feeds them and stay around long. If there’s any hope in overcoming the politicization or the parentage problems, it would come from observing the few orgs that have so far managed it pretty well. Snopes and Wikipedia are two organizations that have managed a pretty good track record. In one case it’s because the company has stayed small and harbors very low commercial and no political aspirations, making it free from some of the pressures typically lead to compromised content. In the other case, it’s through a public battle of personally motivated contributors who counter each other where possible. Would your truth foundation be able to replicate this? Hard to do at scale (and at maximum speed to stay ahead of fake news as it is generated). But I have an idea for this: What if you had a small team of Snopes-like editors who are obsessed with accuracy that identify truth claims and deliver preliminary assessments of likely “truthiness” but you pair them with multiple machine intelligences that are continuously learning from their human editors what is considered true and what is not. Once developed enough, the machine intelligences can take on the massive task of spotting likely false stories at scale, delivering preliminary assessments of “truthiness” for human peers to review. It’s basically taking today’s human+machine chess world and applying it to news. Wouldn’t AI vendors be thrilled to let their machine intelligences compete with each other to be seen as the better truth detectors? As long as the human editors and the machines share their thought processes (formal criteria, preferred sources, etc.) publicly for external check and confidence, we could actually build a very strong base of truth-confirming outcomes. Fake news first, then the increasing problem of fake science (as in, studies that get published but which are fraudulent) next. The more I think of how necessary this is, I hope you get that phone call…