I’m reading a bird forum where a heated debate has broken out over a photo, supposedly the first-ever sighting of a red-winged blackbird in central Brazil, thousands of miles outside where the species normally lives. Excited birders are treating it as a genuine discovery.

Except the bird in the photo isn’t a red-winged blackbird at all. It’s actually a completely different, common local species called an epaulet oriole. Someone had asked an AI tool to "make the picture look better," and the AI quietly redrew the bird’s wing patches into something that matched a species from an entirely different continent.

So I looked into it. Here’s what I found.

AI investors should get ready for a BIG surprise

Do you hold any of these AI stocks?

Wall Street insider Jason Bodner — the man who called Nvidia at $4.50 — says today’s AI stocks are about to hit a wall.

And a completely different group of AI firms… names Wall Street is starting to ignore… are about to take off.

This has nothing to do with SpaceX…

A new chatbot…

Autonomous robots…

Or anything you’re likely hearing about.

It has to do with a brand-new “light-speed” device turning AI as we know it into “Accelerated AI”…

Making it 100 times faster…

And 100 times more energy efficient — right here, on Earth.

Already, some of the biggest tech investors like Elon Musk, Mark Zuckerberg, Cathie Wood, and Bill Gates are moving money into it.

Just to name a few…

But you won’t hear anything about it in the mainstream news…

In fact, TV pundits spent most of this past year talking about AI worries and its “existential risk” to jobs…

Or arguing whether we’re in an AI bubble and when it would pop…

That’s why most Americans won’t see it coming until it’s too late.

Don’t be one of them…

Because if you’re holding the wrong AI stocks when “Accelerated AI” goes mainstream…

You could spend the next decade just trying to claw back to even…

The next 12 to 24 months could hand you bigger gains than the entire AI boom of the last three years.

Click here to hear the full story and get ahead of the crowd.

But hurry, because this opportunity won’t stay hidden much longer.

TODAY’S REALITY CHECK

AI Is Quietly Inventing New Bird Species — And Some Ornithologists Almost Fell For It

610 Million

Photos on iNaturalist

1,400

Flagged for AI So Far

1

False "Country First" Record

Medium

Fooled Index

A small songbird photographed in the wild, its wing markings quietly altered by an AI editing tool into the pattern of an entirely different species

AI-ALTERED — The bird is real. The markings that made it look like a rare species were added by AI.

Millions of people use citizen science platforms like iNaturalist, eBird, and the Macaulay Library to log real wildlife sightings, and scientists lean on that public data to track where species live, how they're moving as the climate shifts, and what's changing in the wild. A new study published in the journal Nature Ecology & Evolution, led by ecologist Dr. Alexander Lees of Manchester Metropolitan University, warns that AI tools are quietly poisoning that data.

The most striking case involved a photo submitted as a possible first sighting of a red-winged blackbird in central Brazil, a species that normally lives thousands of miles away in North America. The bird in the photo was actually an epaulet oriole, a species common to the region. The photographer had simply asked an AI editing tool to sharpen or improve the image, and the AI added red wing patches that weren't there, quietly reconstructing the bird to resemble the wrong species entirely.

Lees and his co-authors, including Tony Iwane of iNaturalist, also demonstrated how easily a species could be invented from nothing at all. Using Google's Gemini, the researchers generated a convincing but entirely fictional image resembling a willow tit, close enough that Lees said it was "almost convincing enough to fool some 'ornithologists.'" The AI-fabricated bird had subtly deformed feet and beak details, but only on close inspection.

The scale of the problem is still coming into focus. Of the more than 610 million images on iNaturalist, only about 1,400 have been flagged for AI use so far, though researchers say that number likely undercounts the real total, since the sneakiest cases are edits to real photos rather than fully invented ones. As Lees put it, "nobody is falling for a toucan sighting in Siberia," but subtle "beautifying" edits are far harder to catch and just as capable of corrupting a scientific record.

🐘

The wing patches were added by AI. The Brazil "red-winged blackbird" was really a common local oriole; an AI editing tool added red patches that were never actually on the bird.

🦃

A whole fictional bird almost passed. Researchers generated a convincing but entirely fake bird resembling a willow tit, close enough to nearly fool experienced ornithologists.

📊

The flagged number is likely too low. Only about 1,400 of iNaturalist's 610 million images have been flagged for AI so far, but researchers believe quiet edits are undercounted.

🔬

Experts, not just casual users, can be fooled. Subtle AI enhancements, not obvious fakes, are the real threat, since they can trick both trained scientists and automated species-ID tools.

FAKE

🚨 AI-Altered, Not a Real Discovery

No red-winged blackbird has actually turned up in Brazil. The "discovery" came from an AI tool quietly redrawing a common local bird into a rare one when a photographer just wanted a sharper picture. Scientists warn this kind of quiet AI editing, not obvious hoaxes, is the real threat to the wildlife data researchers depend on.

WHAT ELSE GOT FLAGGED

A New Book Warning About AI and Truth Was Caught Using AI-Fabricated Quotes

FAKE
The cover of a new book about truth in the age of artificial intelligence, later found to contain quotes fabricated by the very technology it warns about

"The Future of Truth: How AI Reshapes Reality," a book by filmmaker and author Steven Rosenbaum released in May 2026 with a foreword by Nobel laureate Maria Ressa, set out to examine how artificial intelligence is reshaping trust and reality. A New York Times review found more than half a dozen quotes in the book that were either misattributed or entirely fabricated. One fake quote falsely attributed to journalist Kara Swisher didn't sound like her at all; Swisher told the Times she "never said that," adding it made her "sound like I have a stick up my butt, according to ChatGPT." Rosenbaum acknowledged using ChatGPT and Claude during his research and writing process and called the errors "improperly attributed or synthetic quotes," taking full responsibility while noting the incident inadvertently proved the book's own warning about AI and truth.

YOUR REALITY DETECTOR TOOLKIT

Today’s Skill: Ask If ‘Better’ Might Mean ‘Fake’

Any photo that’s been "enhanced" deserves a second look before you trust its details.

01

Ask if a ‘better’ version might mean a fake one

The Brazil bird photo only became fake once someone asked AI to improve it. Any photo that's been "enhanced" by AI deserves a second look before you trust the details.

02

Notice when a claim is a big departure from the norm

A red-winged blackbird in Brazil would be thousands of miles outside its known range. Extraordinary location claims deserve extra scrutiny, not extra excitement.

03

Remember that even experts can be fooled by subtle edits

Trained ornithologists nearly missed the fake willow tit. If professionals can be fooled by careful AI work, a quick glance from anyone else isn't a reliable test either.

“Trust, but verify.”

— Ronald Reagan