Some "facts" refuse to die. They get repeated so often they feel true, and then they get used to sell you something, scare you, or talk you into a bad decision. Each entry here is one of them, run through the same three steps: the belief, the dated receipt from a real source, and the verdict. When we find we're wrong about one, we correct it here, dated.
"It takes 21 days to form a habit."
The most-cited study on this (Lally and colleagues, 2010, in the European Journal of Social Psychology) followed people forming new habits and found it took a median of about 66 days for a behavior to become automatic, ranging from 18 to 254 days depending on the person and the habit. The "21 days" number traces to a 1960 book by surgeon Maxwell Maltz, who observed patients took "a minimum of about 21 days" to adjust. The floor became a rule through decades of repetition.
Verdict: FALSESources: Lally et al., 2010; Scientific American; James Clear summary.
"You only use 10% of your brain."
Brain-imaging research shows that virtually all of the brain is active over the course of a day, and there is no silent 90 percent waiting to be unlocked. Damage to almost any part of the brain has effects, which would not be true if most of it did nothing. Neurologists and major science outlets have debunked this repeatedly.
Verdict: FALSESource: brain-imaging neuroscience, as summarized by Scientific American.
"Big institutions like BlackRock are buying up all the single-family homes."
Government and academic analyses find that small investors, the mom-and-pop landlords, make up the large majority of investor home purchases, while the giant institutional funds own only a small share of single-family homes nationally. Investor activity is real and worth watching, but "Wall Street is buying every house" is not what the data shows. This one matters because the panic version gets used to sell courses and drive bad decisions.
Verdict: OVERSTATEDSources: U.S. GAO on institutional ownership of single-family rentals; Local Housing Solutions.
"Bitcoin only goes up. It's a safe store of value."
Bitcoin has had at least four peak-to-trough crashes of roughly 77 to 93 percent: about 93 percent in 2011, 86 percent in 2015, 84 percent in 2018, and 77 percent in 2022. It has recovered before, but "only goes up" and "safe" are not accurate descriptions of an asset that has lost most of its value four separate times. Understanding the volatility is the point, not talking you in or out of it.
Verdict: FALSESource: Bitcoin price history (major drawdowns, 2011–2022). Educational only, not investment advice.
"Tokenization will be a $30 trillion market by 2030."
That is a forecast, not a fact, and the forecasts are all over the map, from a few trillion to tens of trillions depending on who is selling the report. A projection with a range that wide is a guess dressed as a certainty, and none of it has happened yet. Tokenization is real and growing, but a specific 2030 dollar number stated as fact is exactly the kind of hype a receipts-first reader should flag.
Verdict: UNVERIFIED FORECASTSource: analyst projections vary widely; treat any single 2030 figure as a forecast, dated to whoever issued it. Educational only.
"An MIT study proved 95% of AI projects fail."
A 2025 MIT report did find that most enterprise generative-AI pilots showed no measurable profit-and-loss impact. But the punchy "95 percent fail" version strips out the context, and the methodology and definition of "fail" have been disputed. The honest way to say it is "one 2025 MIT report found most enterprise AI pilots showed no measurable ROI," not "AI fails 95 percent of the time." A real finding, worn as a bigger claim than it can carry.
Verdict: CONTESTED: needs contextSource: MIT report (2025) on enterprise generative-AI adoption; framing disputed. Verify the exact wording before repeating.