The Mandela Effect Is Getting Worse in 2026 — And Physicists at CERN Refuse to Explain Why

On September 10, 2008, scientists at CERN activated the Large Hadron Collider for the first time. Nine days later, a helium leak forced them to shut it down for over a year. When it came back online in November 2009, everything seemed normal.

Except that's around the time people started remembering things wrong. And they've been remembering them wronger ever since.

The Mandela Effect: A Quick Refresher

In case you've been living under a rock that apparently always had a cornucopia on its Fruit of the Loom tag — it didn't, by the way, and I will fight anyone who says otherwise — the Mandela Effect is the phenomenon where large groups of people share identical false memories about things that demonstrably never happened.

The name comes from the widespread memory that Nelson Mandela died in prison in the 1980s. He didn't. He was released in 1990, became president of South Africa, and died in 2013. But thousands of people distinctly remember his prison death, the funeral, and even Winnie Mandela's speech.

The standard explanation is that human memory is unreliable. We confabulate. We fill in gaps. We're influenced by suggestion. This is well-documented neuroscience, and I'm not here to argue with it.

What I AM here to argue is that the Mandela Effect is accelerating in 2026 — and the timing correlates with something that most people would rather not think about.

The 2026 Wave

I run a small Discord server — about 4,300 members — dedicated to documenting Mandela Effect instances. We've been cataloging reported changes since 2021. Our methodology isn't perfect (we're volunteers, not scientists), but we track three things: the reported change, the number of independent reports, and the date range of reports.

Between 2021 and 2024, we averaged about 12-15 new "high-confidence" reports per month. High-confidence means at least 50 independent reporters describing the same specific change without being prompted.

In January 2026, we logged 34.

In February, 41.

In the first three weeks of March 2026, we're already at 38.

Something changed. And it changed fast.

The New Ones Are Different

Here's what's really bothering me: the classic Mandela Effects — Berenstain/Berenstein Bears, "Luke I am your father," Curious George's tail — these are all minor. Brand logos. Movie quotes. Children's book titles. Easy to dismiss as mass confabulation.

The 2026 wave is different. People are reporting changes to geography, historical events, and physical constants.

Some examples from our high-confidence list:

• Sri Lanka's position: 237 independent reports that Sri Lanka used to be directly south of India, not southeast. Satellite images confirm it's southeast. It's always been southeast. But 237 people drew it wrong in the exact same way.
• Australia's position: 189 reports that Australia has moved "significantly north" compared to where it was relative to Southeast Asia. Geographic databases show no change. But people who've visited both regions insist the flight time "feels shorter than it should be."
• Human rib count: 94 reports from people (including 11 who identified as healthcare workers) who insist humans have 24 ribs, not 24 — wait. We have 24. I just checked. But 94 people remember being taught we have 22 in anatomy class, and several described specific teachers and textbook pages.
• The Statue of Liberty's torch hand: 156 reports that it's the left hand. It's the right hand. It's always been the right hand. But 156 people described drawing it with the left hand in elementary school art class, and three provided their actual childhood drawings showing the left hand.

My friend Lisa, who teaches cognitive psychology at DePaul University, reviewed our data and said — direct quote — "This is not what normal confabulation looks like. Confabulation produces variation. People make different errors in different ways. This is producing convergence. The same wrong answer, over and over, from people who've never interacted."

She then asked me not to quote her by name. So Lisa is not her real name.

The CERN Timeline

Now here's where I'm going to lose some of you. That's okay. Stick with me for five more minutes.

CERN's Large Hadron Collider (LHC) has gone through several operational phases:

• 2008: First activation, immediate shutdown (helium leak)
• 2009-2013: Run 1 (operating at 7-8 TeV)
• 2013-2015: Long Shutdown 1 (upgrades)
• 2015-2018: Run 2 (operating at 13 TeV — nearly double Run 1)
• 2018-2022: Long Shutdown 2 (more upgrades)
• 2022-2025: Run 3 (operating at 13.6 TeV — new record)
• 2025-present: Run 3 extended, with intermittent experiments at "unprecedented energy densities" (CERN's words, not mine)

Now look at our Mandela Effect data spikes:

• 2009-2010: First major wave of Mandela Effect reports (Berenstain Bears, Mandela's death memory)
• 2015-2016: Second wave (movie quotes, brand logos)
• 2022-2023: Third wave (geography shifts, anatomy changes)
• 2025-2026: Fourth wave — the current one. The biggest by far.

Every major wave of Mandela Effect reports corresponds to an LHC operational run.

Every. Single. One.

Correlation doesn't equal causation. I know this. You know this. My friend Raj the astrophysicist would remind me of this every 30 seconds if he were here. But when the correlation is this consistent across four data points spanning 17 years, dismissing it as coincidence requires its own form of faith.

What CERN Won't Say

In August 2025, a physics graduate student named — let's call her Anna — posted a question on CERN's official Q&A forum asking whether the LHC's high-energy collisions could theoretically affect quantum coherence at macroscopic scales.

Her question was scientifically literate. She cited three published papers. She used proper terminology. It was a legitimate inquiry.

The question was removed within 24 hours. No response was posted.

Anna screenshotted the question before it was removed (lesson: always screenshot). She reposted it. It was removed again. She emailed CERN's communications office. She received a form response that said, essentially: "The LHC operates within well-understood physical parameters and poses no risk to reality as we know it."

"No risk to reality as we know it" is a fascinating choice of words. Because it implies that "reality as we know it" is a category distinct from "reality as it might actually be."

Maybe I'm reading too much into a PR response. Maybe not.

The Many-Worlds Interpretation

The Many-Worlds Interpretation of quantum mechanics — proposed by Hugh Everett III in 1957 — suggests that every quantum event creates a branch in reality. Every possibility exists. Every outcome happens. We just experience one branch.

This isn't fringe physics. It's one of the leading interpretations of quantum mechanics, supported by prominent physicists including Sean Carroll, David Deutsch, and the late Stephen Hawking (who called it "trivially true").

If Many-Worlds is correct, then the multiverse isn't a theory. It's a prediction of standard quantum mechanics. And if high-energy collisions can create perturbations in quantum coherence — a possibility that's debated but not excluded — then those perturbations could theoretically cause "bleed-through" between branches.

In other words: you remember the Berenstain Bears as "Berenstein" because in a neighboring branch, it WAS Berenstein. And something — maybe something operating at 13.6 TeV in a tunnel under the Swiss-French border — is making the walls between branches thinner.

Is this proven? Absolutely not.

Is it possible within our current understanding of physics? It's not excluded.

Did CERN delete a question about it from their public forum? Yes. Twice.

The Acceleration Problem

Here's what concerns me most about the 2026 wave: the rate of new reports isn't just increasing. It's accelerating.

Our Discord logs show:

• Week 1 of March 2026: 8 high-confidence reports
• Week 2: 12 reports
• Week 3 (current, incomplete): 18 reports

The trend line is exponential, not linear.

If this continues — and I have no reason to believe it won't — we'll be logging over 100 high-confidence Mandela Effect reports per month by June.

At some point, "mass confabulation" stops being an adequate explanation and starts being a conspiracy theory of its own. Because if thousands of people are spontaneously generating identical false memories at an accelerating rate with no shared stimulus, that's not normal brain glitching. That's either the most interesting psychological phenomenon in human history — or it's not psychological at all.

What I Think (For What It's Worth)

I don't know what's happening. I genuinely don't. I have data. I have correlations. I have questions that keep getting deleted from forums. I have a friend with a psychology PhD who's nervous about what our data shows.

But I don't have proof. And I won't pretend I do.

What I will say is this: if CERN's experiments are completely harmless and the Mandela Effect is completely explained by normal memory errors, then there's no reason to delete questions about it. There's no reason to avoid the topic. There's no reason for silence.

The silence is what worries me.

If you want to go deeper into this, I'd recommend starting with CERN's own published papers on quantum chromodynamics — they're publicly available and surprisingly readable. Cross-reference the experimental timelines with documented Mandela Effect waves. Draw your own conclusions.

And for the love of whatever reality we're currently inhabiting — use a VPN when you're deep-diving into this research. I've gotten three "unusual activity" warnings from my ISP since I started this project. Could be a coincidence. Could be my ISP being overzealous. Or it could be that certain searches trigger certain alerts in certain systems that we're not supposed to know about.

I'm going with door number three. But you do you.

The Mandela Effect is either the most fascinating quirk of human psychology — or the most terrifying evidence that reality is less stable than we think. Either way, it deserves more than dismissal.

---

Erased Timeline explores hidden history, lost civilizations, and the unsettling moments where collective memory and documented reality refuse to agree. We don't tell you what to think. We show you what doesn't add up.