Yesterday scientists published proof that AI has a fundamental cognitive weakness.
Not on a hard problem. Not on a complex benchmark.
On a test designed for undergraduate psychology students.
The paper was published June 10, 2026 in PNAS Nexus, one of the most rigorous scientific journals in the world. Three researchers from Texas Tech University and the City University of New York. One finding that reframes every confident claim about AI capability you have heard in the last two years.
Here is the test they used.
The Stroop task. Invented in 1935. Used in psychology labs for 90 years. You see a word, the word says RED but the ink is printed in blue. Your job is to name the ink color. Not read the word. Just name the color.
Your brain fights itself. The obvious answer, RED is wrong. The correct answer, blue requires suppressing the thing your mind wants to do automatically.
That suppression is called executive control. It is one of the most fundamental measures of cognitive function humans possess. It underlies everything important following complex instructions, maintaining a rule across a long task, catching when a later piece of information contradicts an earlier one, noticing when you are about to give the wrong answer because the obvious answer is wrong.
Researchers gave top AI models the classic attention test and found a major flaw. While the models could correctly name colors in short lists, their performance deteriorated sharply as the task became longer and more complex.
Short list. The AI is fine. Gets it right. Looks capable. Impressive even.
Longer list. Performance collapses.
The paper describes the finding as deficient executive control in transformer attention.
Not slower. Not less accurate. Deficient.
Here is what makes this alarming beyond the benchmark score.
Every enterprise AI deployment in the world right now is built on an assumption. The assumption is that if AI performs well on the demonstration the controlled test, the curated benchmark, the polished proof of concept, it will perform comparably on the real task.
The Stroop finding breaks that assumption at its foundation.
Short task. Looks fine.
Long task. Deficient.
The demo is always short. The real work is always long.
A legal AI reviewing a 300-page contract needs to maintain a rule flag this clause type,across hundreds of pages of text. A medical AI analyzing a complex patient history needs to hold context across dozens of symptoms, test results, and medications without letting the obvious pattern override the correct one. A financial AI auditing a large dataset needs to catch the exception buried in page 47 of a 60-page report.
These are Stroop tasks. They are tasks where the obvious answer is wrong and executive control is the only thing that catches it.
And the paper published yesterday says that executive control in transformer-based AI models is deficient under exactly the conditions where it matters most.
The AI does not know the task is getting harder. It does not experience cognitive load increasing. It does not know it is failing. It generates the wrong answer in the same confident tone it used for all the right ones.
There is no uncertainty signal. No "this is getting complicated." No slowdown that would prompt a human to pause and check. Just a confident answer that is increasingly wrong as the task grows.
AI passed the bar exam. Scored 90% on elite mathematical competitions. Achieved human-level performance on medical licensing exams. Every benchmark the AI industry uses to demonstrate capability is a short Stroop task.
The real work is the long one.
Yesterday researchers published proof that the long one is where AI breaks.
Source: Patel, Wang, Fan · Texas Tech University + City University of New York · "Deficient Executive Control in Transformer Attention" · PNAS Nexus
( Link in the comments)
It’s amazing how much sustainability damage for average folks they’ve been able to pull off in 18 months. Not just a wrecking ball but precision dismantling of protections from sea to shining sea, latest is opening of Cape Cod marine sanctuary to all nationality fishing.
Let's walk through what actually happened here, in order.
DOGE cut the USAID program specifically designed to prevent screwworm from crossing the U.S.-Mexico border. DOGE cut USDA's animal disease control and prevention funding. That funding had supported more than 180 outbreak investigations in 22 countries and capacity-building in more than 160 laboratories. The screwworm monitoring and response program that watched the border for exactly this parasite - cut.
Then screwworm showed up in Texas cattle. Texas Governor Greg Abbott declared a disaster for Zavala and Uvalde counties this week.
Agriculture Secretary Brooke Rollins went on CNBC this morning and blamed the Biden administration, 17 months out of office.
Her specific words: "obviously not much had been done to push back."
The program that was supposed to push back existed. DOGE eliminated it in March 2025. Rollins has been Agriculture Secretary since February 13, 2025. The cuts happened on her watch.
Beef prices are already high. Ranchers in south Texas are now dealing with a flesh-eating parasite that was eradicated in this country in the 1960s - eradicated, specifically, using the sterile fly program her department defunded.
The flies existed. The program existed. The budget existed.
Until it didn't.
Mark this moment.
This is where the endgame begins—though it might take two years to play out.
A President of the United States is openly linking the *Pentagon* and rigging elections for his party.
He is framing an authoritarian takeover as something the *U.S. military* wants.
Do not doubt that if a major pandemic hits, millions of Americans will die because of this grotesque man. And not a single person better say that we didn’t know it was coming. Alarm bells have been ringing *nonstop* that this sick buffoon will kill people. https://t.co/PUXZBHP2h2
A retired lawyer in the U.S. was watching the news when he saw the story about the new Trump commemorative coin and something immediately didn't sit right with him.
So he did what lawyers do. He went digging...
And he found it. A federal law passed in 1866 that explicitly prohibits living people from appearing on U.S. currency. It's not a grey area. It's not open to interpretation. It's been sitting in the books for over 150 years.
The last time this actually happened was 1926 when a coin featuring Calvin Coolidge was minted while he was still alive and serving as president. The backlash was immediate. The coins were pulled. And the law was reaffirmed...
Now this retired lawyer has filed a federal lawsuit against the U.S. Mint not because of who is on the coin, but because the law says it simply cannot be done. Full stop...
No political agenda. No protest. Just one guy, a dusty legal statute, and a federal case that nobody in Washington apparently saw coming
Stages of pro-Israel Trumpist rationalization.
1. Regime change!
2. OK, but at least the nuclear program will be totally ended.
3. OK, but at least the missiles will be gone.
4. OK, but at least sanctions will stay.
5. OK, but at least Israel can finish the job in Lebanon.
Sad.
And as an EXCLAMATION POINT to this ACCURATE and Devastating, truth-telling editorial, there's the @GOP Senate Select Committee on Intelligence. THEY were NOT restricted in scope or possible legal outcomes as Mueller was by Trump. 1/
https://t.co/HHlJS44MGY
The USS Abraham Lincoln is one of the most powerful weapons ever built. The Americans built it to park wherever they wanted and dare anyone to object. Iran objected.
The Lincoln started this war sitting 350 kilometres off the Iranian coast. Close enough to smell the place. Then something happened. Nobody agrees on what. Trump says 101 missiles were fired and every single one was knocked down. Iran says they hit it. CENTCOM says nothing came close. Everyone is lying about something, and the ship is now 1,100 kilometres away, tucked behind Oman’s coastal mountains like a man hiding behind a sofa.
That distance matters more than any press release. A carrier at 350 kilometres is terrifying. Its jets can reach Tehran in minutes. Its strike package lands before anyone has time to make a phone call. A carrier at 1,100 kilometres is a different proposition entirely. Longer missions. More tankers. More exposure. More things that can go wrong on the way there and the way back. You don’t move a ninety-thousand-ton symbol of national power three times further from the fight because the laundry room caught fire.
Which brings us to the USS Gerald R. Ford. The world’s largest aircraft carrier caught fire in the Red Sea on March 12. The blaze started in the aft laundry facility, caused a major damage control response, displaced sailors across the ship, and destroyed over 100 beds.  The Navy flew a mattress airlift from a carrier still sitting in Virginia. They also collected nearly 2,000 sweatsuits to distribute to the crew because most of the laundry was out of commission.  Then the Ford left the war zone entirely and sailed to Crete for repairs. 
But the fire was almost a relief from the Ford’s other problem. The $13 billion ship’s vacuum sewage system, borrowed from the cruise ship industry, has been failing throughout the deployment. Out of nearly 650 toilets onboard, most have been non-functional at various points. Sailors have waited up to 45 minutes in line.  The maintenance crew was working 19-hour days trying to keep up with the demand. Everything from t-shirts to a four-foot piece of rope has been pulled from the pipes.  The most powerful warship ever built. Broken toilets.
So here is the full picture.
The Abraham Lincoln, pushed 750 kilometres south by something Iran apparently fired at it, is now launching aircraft on extended missions from behind a mountain range in Oman. The Gerald R. Ford, the carrier sent to replace it as the tip of the spear, caught fire and sailed to Greece. And somewhere in a Virginia shipyard, the USS George H.W. Bush is heading out to join a war that has already humiliated the first two ships sent to fight it.
This is the part the Pentagon doesn’t want to discuss at press conferences. Modern precision missiles, fired in large enough numbers, are doing something that was supposed to be impossible: making American carriers think twice about where they stand. If the Lincoln was actually struck, even once, the implications reach far beyond one ship and one war. They reach into every strategic calculation every American adversary has been running for thirty years.
The United States built its entire post-Cold War foreign policy around the ability to park a carrier strike group off your shore and explain, very politely, that the conversation was now over. That era may not be over. But for the first time in a generation, someone is making it complicated. And the Lincoln, sitting 1,100 kilometres from the action it was sent to dominate, is the clearest evidence yet that something has changed.
Nobody builds a carrier that big just to hide it behind a mountain range in Oman. And nobody sends a second one to Greece to fix the toilets.
Follow Gandalv @Microinteracti1
The Villages retirement community in Florida have turned out in huge numbers for No Kings. William Steiner, our Villages correspondent, says that the turnout is twice as big as last year's.
I want to flag something that may have flown under the radar, but shouldn't. In December 2025, President Trump and @SecVetAffairs Doug Collins moved to ban access to abortion care at the VA, even in cases of rape, incest, or when the health of the mother is at risk. Earlier this week, the Senate voted on our bill to overturn that policy. 50 Republican Senators opposed it, meaning the abortion ban remains in effect.
This threat to women's healthcare continues to be real. After losing elections in the wake of Roe falling, Republicans know that it's not politically popular to be publicly for abortion bans. So they do it just like this: quietly, in the dark of night, through bureaucratic rule-making -- hoping we won't catch it. Our veterans deserve access to abortion care. Period. And I will not stop shining the light on this and working to overturn this new ban.
https://t.co/Ue5khGufWo
Pressed by my questions, Trump’s judicial nominees give identical, nonsensical canned responses—looking ridiculous & demonstrating an abject absence of independence & integrity. Lacking a backbone now, they won’t have one on the bench.
1. Unpopular war.
2. Unpopular tariffs.
3. Unpopular TSA shutdown.
4. Unpopular corruption.
5. Unpopular health care cuts.
6. Unpopular 4.2% inflation.
6. Oh...and saving the best for last: a deeply unpopular President!
Republicans have so much to run on this Fall!