Empirical Earth · August 12, 2026

Something real is happening.
It is not a gravity loss.

On August 12, 2026, the Moon’s shadow crosses Iceland and Spain. Here is the truth behind the headline you were sent.

days
hours
min
sec
to greatest eclipse · 17:45:54 UTC, off western Iceland
The claim you saw

“Earth loses gravity for 7 seconds”

“On August 12, 2026, the world will lose gravity for 7 seconds. NASA knows. A leaked document, Project Anchor, an $89 billion budget, 40 million deaths from falls, at 14:33 UTC.”

The date is real. There genuinely is a total solar eclipse on 12 August 2026, and it genuinely is the first over mainland Europe since 1999. Anyone checking that part will find it holds up. The rest is not. There is no NASA document, project, or budget called Project Anchor. Fact-checkers found no trace of the name before the first viral post on December 31, 2025, and NASA has since said so on the record: the agency told fact-checkers plainly that Earth will not lose gravity on August 12, because total gravitational force is set by the mass of the Earth system and nothing that day changes it. And the physics is simple: gravity comes from mass, not from shadows. A solar eclipse is only the Moon’s shadow sweeping across the ground. It changes nothing about the Earth’s mass, so it changes nothing about your weight. You will not float. Nobody will.

Two details are worth keeping. The account that posted the original claim went offline in the first week of January 2026, before the first fact-check was even published — so the story has been circulating for months without the source it cites. And the budget has drifted in retelling, from $89 billion in the original post to $89 million downstream. A leaked document does not change its own numbers on the way through a crowd.

The hoax worked because it stapled a fake catastrophe to a real date. The date is real. The catastrophe is invented. Once you separate the two, what is left is genuinely worth seeing.

The test

Two predictions. One of them is checkable on the day.

Both stories name a time. That makes this rare: you do not have to take anybody’s word for it, and you do not have to wait long.

The hoax says the event happens at 14:33 UTC, and that gravity fails for seven seconds.

Astronomers say greatest eclipse falls at 17:45:54 UTC on the clock (catalogues print this as 17:47:06 in Terrestrial Dynamical Time, which runs about 73 seconds ahead of clock time), at 65°12′ N 25°12′ W, about 45 km off western Iceland, with totality lasting 2 minutes 18 seconds at the center line. Those numbers come from orbital mechanics and were computed decades ago. The two times are three hours and thirteen minutes apart, so on the day, one of them is simply wrong.

This eclipse is number 48 of 72 in a family called Saros 126, a repeating pattern spaced 18 years and 11 days apart that has been tracked since antiquity. Its place in that sequence, its magnitude of 1.0386, and the fact that the Moon passes perigee 2.2 days beforehand and so appears slightly larger, all come from the same geometry that puts the shadow on Iceland at 17:45:54. None of it was adjusted after the hoax appeared. The predictions were published long before December 2025.

So watch the clock. If the shadow arrives where and when the orbital mechanics say it will, the model that produced that number is doing something the hoax cannot do. And if nobody floats at 14:33, that is worth remembering the next time a leaked document turns up with a number attached.

The real event

A total solar eclipse over Europe

For a little over two minutes, the Moon covers the Sun completely, and its corona — the Sun’s outer atmosphere — comes into view. The last time this happened over these places, your great-grandparents may have watched it.

2m 18s
longest totality, 45 km off western Iceland
1954
last total eclipse seen from Iceland
1905
last total eclipse seen from Spain
~100/hr
Perseid meteors peak the same night
The path

Where the shadow goes

This eclipse is unusual: it runs east–to–west across the top of the world, arcing over the Arctic and just missing the North Pole before dropping south.

That backward, over-the-pole path is not a quirk of storytelling. It is exactly what the geometry of a tilted, spinning globe and an orbiting Moon requires — and it is one of the things a flat map cannot reproduce.

If you can see it

Watch it safely

Only during the brief moments of totality is it safe to look without protection. Before and after, and everywhere outside the narrow path where the eclipse is only partial, you must use eclipse glasses that meet the ISO 12312-2 standard. Regular sunglasses are not safe, no matter how dark. Never look at the Sun through a camera, telescope, or binoculars while wearing eclipse glasses.

Want to know how we know?

What would show this page is wrong: the Moon’s shadow arriving somewhere other than the published path, or at a time other than the published one; or any measurable change in weight anywhere on Earth during the eclipse. Both are checkable on 12 August by anyone with a clock.

Empirical Earth takes every flat-Earth and eclipse claim, states it at its strongest, then tests it against evidence you can check yourself.