Ask an AI chatbot a creepy question and you’ll usually get a creepy answer. This weekend, an X user named Adrián Sanpei tagged Elon Musk’s chatbot with a simple prompt: “Hey @grok, what’s the scariest thing in the universe that almost no one knows about?”
Grok didn’t go for black holes, killer asteroids or aliens. It went for something much stranger. “The universe could be in a metastable state,” it replied. “At any moment, anywhere, a bubble of true vacuum could randomly appear, expand at the speed of light, and rewrite the laws of physics, destroying all matter without any warning.”
Sanpei’s response was perfect: “Thanks for making my anxiety worse grok.” As UNILAD Tech noted, the thread quickly filled up with people suggesting other terrifying things, from the deep ocean to the idea of infinity. One reply summed up the mood: “That’s probably the least frightening way to go. Instantaneously.”
So is Grok right? Mostly, yes. False vacuum decay is a real idea in physics, and it’s been studied seriously for about 50 years. But the chatbot left out the part that matters most: how unlikely it is. Here’s the full picture.
The short version
- False vacuum decay is a real, peer-reviewed idea: our universe might not be in its most stable possible state.
- If a bubble of “true vacuum” formed, it would spread at nearly the speed of light and change the physics inside it. Atoms as we know them couldn’t exist.
- Measurements of the Higgs boson and top quark suggest our universe really is metastable, but the math isn’t settled.
- The best estimates put the expected wait at around 10161 years. The universe is about 13.8 billion (roughly 1010) years old.
- Grok’s answer was accurate but dramatic. It skipped the odds.
What is a “false vacuum,” anyway?
In physics, “vacuum” doesn’t mean empty space the way we normally think of it. It means the lowest energy state that space can settle into. Every particle and force we know about depends on the properties of that state. Change the vacuum, and you change the rules.
The easiest way to picture it is a ball rolling on a hillside. The ball has come to rest in a little dip. It looks stable, and if you nudge it slightly, it rolls back. But further down the hill there’s a much deeper valley. The little dip is the false vacuum. The deep valley is the true vacuum. The worry is that our universe is the ball sitting in the dip.
The “ball on a hill” picture of vacuum decay
Energy of the vacuum (simplified illustration)
Here’s the twist. In classical physics, a ball in a dip stays there forever unless something kicks it over the hump. In quantum physics, it doesn’t need a kick. Particles can “tunnel” through barriers they shouldn’t have enough energy to cross. It’s the same effect that makes nuclear fusion in the Sun possible. Apply that to the whole vacuum, and there’s a tiny chance that somewhere, some day, a patch of space simply tunnels into the true vacuum on its own.
If you like this kind of brain-bending physics, our explainer on why the quantum vacuum isn’t really “nothing” is a good companion read.
What would actually happen
Grok’s description was roughly right. A bubble of true vacuum would form somewhere and start expanding at very close to the speed of light. Inside it, the basic constants of nature would be different. Particles might have different masses, forces might behave differently, and the chemistry that holds atoms and molecules together would likely stop working.
Because the bubble’s wall would move at nearly light speed, there would be no warning at all. Light from the event couldn’t outrun the bubble by any meaningful margin, so nobody would see it coming. That’s the part that makes people uneasy. It’s also why one X reply called it the least frightening way to go: you’d never know it happened.
A detail Grok didn’t mention
The universe is expanding, and that expansion is speeding up. A bubble that formed far enough away could never reach us, because the space between us would grow faster than the bubble could cross it. So even if vacuum decay happened somewhere, it would have to start in our cosmic neighborhood to matter to us.
Where the idea came from
This isn’t a TikTok theory or a sci-fi plot. It’s been in the physics literature since the 1970s.
Fifty years of false vacuum research
Key milestones
1974
Soviet physicists Igor Kobzarev, Lev Okun and Mikhail Voloshin publish one of the first studies of how a false vacuum could decay through bubbles.
1977
Harvard’s Sidney Coleman publishes “Fate of the False Vacuum,” the paper that gave the field its math and its name.
1980
Coleman and Frank De Luccia add gravity to the picture.
2012
CERN discovers the Higgs boson at about 125 GeV. That mass, combined with the heavy top quark, points toward a metastable universe.
2018
Harvard physicists Anders Andreassen, William Frost and Matthew Schwartz calculate the Standard Model’s vacuum lifetime: about 10161 years, and at least 1065 years with 95% confidence.
2024
Researchers in Italy watch “false vacuum” bubbles form in a supercooled atomic gas, a lab stand-in for the real thing.
2025
A team from the University of Leeds, Forschungszentrum Jülich and ISTA simulates bubble formation on a 5,564-qubit D-Wave quantum annealer, published in Nature Physics.
Why the Higgs boson matters here
Whether our vacuum is stable depends on the shape of the Higgs field’s energy landscape, and that shape depends on the masses of particles that interact with it. The two numbers that matter most are the Higgs mass, around 125 GeV, and the top quark mass, around 172 to 173 GeV. Plug them into the Standard Model and the result lands in a narrow zone physicists call “metastable.” Not doomed, not perfectly safe, just sitting in a dip.
The catch is that the answer is extremely sensitive to the top quark’s exact mass. A small shift in that measurement could move us into the fully stable zone. It’s also possible that new physics beyond the Standard Model changes the picture entirely, for better or worse. That’s why scientists describe metastability as the current best reading, not a final verdict.
How long do we have?
This is where Grok’s answer got a little too spooky. “At any moment” is technically true, in the same way you could technically win the lottery a thousand times in a row. The calculated timescales are almost impossible to picture, so here’s a comparison using powers of ten.
Cosmic timescales, compared
Years, shown as powers of ten. Bar length is the exponent, so each step is 10 times longer, not just a bit longer.
Age of the universe today: ~1010
Last new stars form: ~1014
Minimum proton lifetime (experimental limit): ~1034
Vacuum lifetime, 95% lower bound: ~1065
Biggest black holes evaporate: ~10100
Vacuum lifetime, central estimate: ~10161
Put another way: the universe has existed for about 1010 years. Even the cautious lower bound of 1065 years is 1055 times longer than that. By the time vacuum decay becomes a realistic concern, the stars will be long gone and even the largest black holes may have evaporated.
| What Grok said | What the science says |
|---|---|
| “The universe could be in a metastable state” | Accurate. Current measurements point that way, though it isn’t proven. |
| “At any moment, anywhere” | Technically true, misleading in practice. The odds of it happening in our lifetime are absurdly small. |
| “Expand at the speed of light” | Close. The bubble wall would approach light speed. |
| “Rewrite the laws of physics” | Fair. Particle properties and forces inside the bubble would be different. |
| “Almost no one knows about it” | Not really. It’s a staple of physics lectures and pop science videos. |
Can scientists actually test this?
Not directly. Nobody is going to make a bubble of true vacuum in a lab, and that’s a good thing. What researchers can do is build systems that behave in a similar way and watch how bubbles form. In 2024, a team in Trento, Italy, saw bubbles appear in a supercooled gas of atoms that was set up to mimic a false vacuum. In early 2025, physicists used a D-Wave quantum annealer with 5,564 qubits to simulate the process. They found that bubble formation is a collective effect, with bubbles interacting and influencing each other rather than popping up in isolation.
The Leeds-led team was clear that this was a simplified one-dimensional model, not a window into our actual universe. But experiments like these help test the math physicists rely on, which is the only way we can check those 10161-year estimates.
Why this went viral now
Part of it is Grok itself. Musk’s chatbot has a reputation for blunt, slightly edgy answers, and people love posting screenshots of it saying something unsettling. It’s the same AI that’s been busy solving cryptic codes on X and that got a big coding upgrade with Grok 4.6 earlier this year. When a chatbot casually tells millions of people the universe could end without warning, that travels fast.
The other part is timing. 2026 has been heavy on doomsday headlines, from blackout warnings to super El Niño forecasts. A cosmic threat that sounds unstoppable lands right in that mood.
The takeaway
Grok picked a genuinely fascinating piece of physics and described it pretty well. What it didn’t do was give the odds, and the odds are the whole story. False vacuum decay is possible in principle, unproven in practice, and expected on a timescale so long that the universe will be cold and dark well before it becomes a real risk. It’s a great thing to think about. It’s not a great thing to lose sleep over.
About this article: GeekBlog covers U.S. technology news, AI, phones, smartwatches and gaming. Every story is written and checked under our Editorial Policy. Spotted a mistake or have a story tip? Contact our editors.

