Authored by Maryanne Demasi via The Brownstone Institute,
For months, a quiet battle has been unfolding inside the US Food and Drug Administration (FDA).
It began with an analysis of child deaths after Covid vaccination, followed by strategic leaks to major media outlets, and has now erupted into the open with a memo from the regulator’s own vaccine chief.
In September, it was reported that FDA officials had privately investigated 25 paediatric deaths following Covid vaccination — the first systematic review of such cases since the rollout began.
The findings were meant to be presented to the CDC’s Advisory Committee on Immunization Practices (ACIP). But the presentation never came. The meeting passed without a word. Something had happened behind closed doors.
Now we know what.
On 13 November 2025, STAT published an extraordinary insider account describing a tense internal meeting in which FDA scientist Dr Tracy Beth Høeg presented evidence of young people who had died after Covid vaccination.
According to STAT, her findings triggered pushback from career FDA regulators who feared the implications of acknowledging fatal cases.
Now, comes the explosive memo from FDA vaccine chief Dr Vinay Prasad, confirming — for the first time — that US regulators have formally attributed at least 10 of these children’s deaths to Covid vaccination.
Prasad called it “a profound revelation” with far-reaching implications for American vaccine policy, adding that the true number is “certainly an underestimate.”
Here, I’ll take you through the memo, the leaks, the internal rebellion at FDA, and what this means — not just for Covid vaccines, but for all vaccine approvals going forward.
This story marks a turning point in US vaccine regulation.
The Story That Divided the RegulatorIn early September, insiders at the FDA and CDC quietly told the New York Times and the Washington Post that the agency had begun investigating child deaths reported to VAERS.
My reporting confirmed that Dr Tracy Beth Høeg, a senior adviser within the FDA’s vaccine division, had led the review — contacting families, gathering medical records, and obtaining autopsy findings.
click image for story
It was the first case-by-case evaluation of paediatric deaths conducted since the vaccines were authorised.
The review identified twenty-five children whose deaths occurred following vaccination. Those findings were expected to be presented to ACIP on 18–19 September. Instead, without explanation, the discussion disappeared from the agenda.
Even FDA Commissioner Dr Marty Makary had hinted at the findings on CNN, saying, “We’ve been looking into the VAERS database self-reports, [and] there have been children that have died from the Covid vaccine.”
He described an “intense” investigation involving doctors, autopsies, and family interviews. Yet ACIP heard nothing.
Had the FDA reversed course — or had internal forces blocked disclosure?
STAT’s reporting offered the first real clues.
Inside the FDA: The Meeting That Changed EverythingSTAT described a confidential gathering of FDA vaccine scientists in which Høeg presented slides listing roughly two dozen deaths of young people following vaccination.
One slide reportedly read: “Timing fits. Diagnosis fits. No better explanation found. Sufficient information provided.”
According to STAT, some career regulators reacted with “quiet horror” — not at the deaths themselves, but at the policy implications of acknowledging them.
The article portrayed Høeg as pushing to bring the findings to ACIP and to amend vaccine labels for younger males, while longtime staff resisted, describing the evidence as “thin” and worrying about restricting vaccine access.
STAT reported that “no career regulator would stand by the decision,” and Høeg backed away from presenting the cases to ACIP.
It was a rare glimpse of a regulator divided against itself: career staff trying to contain the findings, and FDA leadership apparently trying to surface them.
Nothing more was said publicly — until Prasad’s memo detonated inside the agency.
Prasad’s Explosive MemoThe memo from Dr Vinay Prasad, Director of the FDA’s Centre for Biologics Evaluation and Research (CBER), is unlike anything ever issued by a senior US vaccine regulator.
Addressed to all CBER staff, it confirmed what STAT only implied: FDA scientists had determined that “at least 10 children have died after and because of receiving Covid-19 vaccination.”
Prasad wrote that the true number is “certainly an underestimate” and that “the real number is higher.”
He wrote that “deaths were reported between 2021 and 2024, and ignored for years,” calling it a systemic failure that “requires humility and introspection.”
“It is horrifying to consider that the US vaccine regulation, including our actions, may have harmed more children than we saved,” he wrote.
Prasad defended Høeg’s analysis, saying “Dr Hoeg was correct in her assessment,” and that disagreements reflected subjective coding — not differing facts.
He also noted that healthy children at extremely low risk from Covid had been “coerced” into vaccination under Biden-era mandates, some of which he said “were harmful.”
He added that it was “difficult to read cases where kids aged 7 to 16 may be dead as a result of covid vaccines.”
Prasad also challenged one of the most repeated claims in pandemic messaging — that Covid infection causes more myocarditis than vaccination.
He argued that these comparisons rely on faulty denominators, because they count only people sick enough to seek hospital care while ignoring the far larger number of infections that never present to clinics.
He underscored that vaccination does not prevent eventual infection, so the comparison cannot be framed as “virus versus vaccine.”
A vaccinated child still encounters the virus over their lifetime — but now carries the additional myocarditis risk from the vaccine itself.
The LeaksPrasad’s memo contained another revelation — confirmation of internal sabotage inside the FDA.
He wrote that “slides she presented, emails she sent, and distorted firsthand reports” from Høeg’s meeting had been leaked to media outlets by staff who believed they were acting appropriately.
He condemned the behaviour as “unethical, illegal, and…factually incorrect,” a blunt repudiation of how the STAT narrative had framed events.
In Prasad’s telling, Høeg had not exaggerated the evidence at all. She had uncovered what the FDA had failed to recognise for nearly three years — that Covid vaccines had killed children.
Far from being the rogue figure depicted in selective leaks, she was doing precisely what the public assumes a regulator does: investigating deaths, contacting families, gathering records, and treating each case as a potential signal that demands scrutiny.
For Prasad, the leaks weren’t merely improper — they betrayed the core obligation of a scientific agency.
He said internal debates must remain inside the FDA until ready for public release, and that he would not “endorse selective reporting of our meetings and documents.” Anyone unwilling to follow that principle, he said, should resign.
It was an extraordinary directive — and a clear sign that the internal battle over whether to acknowledge children’s deaths had reached a breaking point.
A Reaction from Inside ACIPWhen the memo surfaced, ACIP vice-chair Dr Robert Malone issued his own statement.
He wrote that he had been aware of the review through ACIP’s internal working group, and that the child deaths “have been known since this summer but not released to the public due to the need to validate the initial findings independently.”
Bound by confidentiality, he could only say, “I have seen the data and findings, and they are even more stunning than this strongly worded letter indicates.”
He said he was “stunned, gobsmacked,” adding: “The significance and importance of this letter in the context of US and global vaccine policy cannot be overestimated. This is a revolution, the likes of which I never expected to see in my lifetime.”
Malone then took aim at the Covid-19 mRNA products: “These products do not work. They do not prevent disease and death. And as Secretary Kennedy testified in the Senate, objective analysis cannot even demonstrate that, on balance, they saved lives.”
MIT professor Retsef Levi — who leads ACIP’s Covid-19 Vaccines Workgroup — issued a similarly forceful response.
He wrote, “the acknowledgement that at least 10 children died from COVID vaccination must be followed with disclosure to the parents,” and said regulators and media “have gaslighted the vaccine injured, including the parents who lost their precious child.”
He described disclosure as “a moral imperative” and essential for any hope of trustworthy vaccine programs.
Inside ACIP, the memo is being understood not only as a scientific shift — but an ethical reckoning.
Critics Rise UpPredictably, the memo triggered pushback from establishment figures who have spent years defending the Covid vaccines from scrutiny.
Dr Paul Offit — a long-time industry-aligned vaccine promoter and a familiar voice deployed whenever safety concerns arise — dismissed the memo as “science by press-release.”
He argued that the memo lacked context and should not be treated as evidence, calling the memo “irresponsible” and “dangerous.”
But Prasad’s communication was never presented as a scientific publication. It was an internal memo to staff. Offit’s attempt to judge it by academic-paper standards is a tactic to avoid addressing what the memo actually says — that children died and regulators overlooked it.
Former CBER director Dr Peter Marks — whose tenure is explicitly criticised in the memo for failing to identify child deaths for years — said he was “taken aback by the clearly political tone of the communication.”
But Prasad’s memo details precisely why Marks’s era is under scrutiny, including his 2021 decision to push out senior FDA officials Marion Gruber and Philip Krause after they objected to the Biden administration’s rush toward booster approval.
If anything was political, it was that episode.
For years, figures like Offit and Marks insisted that VAERS was a robust early-warning system — and that anyone citing it without follow-up investigation “didn’t understand pharmacovigilance.”
Now that FDA investigators have actually done the follow-up — contacting families, obtaining medical records, and reviewing autopsies — these same voices suddenly claim VAERS can’t establish causality at all.
This is the core hypocrisy. You cannot praise VAERS as the backbone of vaccine safety, then declare its signals meaningless once they are properly investigated.
Critics also warned that stricter evidence requirements — such as randomised trials and rejection of surrogate endpoints — would “slow innovation” or “harm vaccine confidence.”
But vaccine confidence is already shattered. Fewer than 10% of American healthcare workers took last season’s Covid booster.
Trust collapsed not because regulators asked too many questions — but because they asked too few, dismissed safety concerns that later proved real, and insisted on messaging long after the data had shifted.
The problem for these critics is not that children have died after vaccination. The problem is that the regulators have finally acknowledged it.
The Future of Vaccine Regulation in the United StatesPrasad’s memo goes far beyond confirming child deaths. It announces a structural overhaul of vaccine oversight.
He wrote that future vaccine approvals would require randomised trials for most new products; that immunogenicity studies would no longer be accepted as proof of effectiveness in new populations; and that vaccines for pregnant women would not be authorised on unproven surrogate markers.
He committed to rewriting the US influenza vaccine framework and overhauling assessments of concomitant vaccination.
Most strikingly, he declared that vaccines would be treated as “no better or worse” than any other medical product — ending decades of special regulatory leniency.
“Never again,” he wrote, “will the US FDA commissioner have to himself find deaths in children for staff to identify it.”
A Global Shift BeginsThe ACIP meeting on 4–5 December will be the first held under these new realities — with the knowledge that the FDA has attributed paediatric deaths to Covid vaccination, that senior leadership has repudiated the previous regulatory approach, and that a revolution in evidentiary standards is underway.
Because many international regulators track the FDA, the acknowledgment that children died from the Covid-19 vaccine — and that the agency failed to detect it — marks a seismic moment in global vaccine policy.
For bereaved families, the acknowledgment is devastating but necessary. For the public, it signals that the institutional silence of the pandemic era is beginning to fracture.
The reckoning has begun.
Republished from the author’s Substack
Tyler Durden Tue, 04/21/2026 - 17:40For years, Democrats and the mainstream media treated 2020 as settled history: the system worked, the election was secure, and accusations of fraud were conspiracy theories.
However, a newly declassified intelligence memo, paired with fresh whistleblower allegations, points in a less convenient direction.
Behind the scenes, U.S. intelligence warned well before the 2020 election that core election systems were more exposed than the public was told, especially the vast digital repositories that hold voter registration data. Making matters worse, according to former senior cyber official Christopher Porter, intelligence leaders then kept those warnings from public view because airing them could have benefited President Donald Trump and complicated the push to portray Joe Biden’s eventual victory as unquestionable.
On January 15, 2020, the National Intelligence Council (NIC) produced an assessment warning that foreign adversaries could compromise U.S. election infrastructure in the coming presidential election, which has just been declassified. The memo specifically called out Russia, China, Iran, North Korea, and other non-state actors. Analysts did not claim they had evidence of a specific plot to alter votes nationwide, but they did say the threat was real, technically plausible, and serious enough that senior intelligence officials personally briefed President Trump at the White House in February 2020.
What worried analysts most was not some Hollywood-style rewrite of every ballot cast in America. “We assess that centralized election-related data repositories, such as voter registration databases, pollbooks, and official election websites, are most vulnerable to exploitation, and adversaries could use access to these systems to disrupt election processes,” the NIC assessment warned.
Intelligence analysts believed vote tabulators and reporting systems had weaknesses, especially machines without paper backups. Despite this, they judged it would be hard for foreign adversaries to change the certified national outcome through direct machine compromise alone. That was never the same as saying the systems were secure in any ordinary sense. It meant large-scale outcome manipulation looked difficult, while localized disruption and perception management looked much easier.
Despite the warnings of threats, after the election, senior officials pushed the opposite narrative, assuring Americans that 2020 had been a model of resilience.
In mid-November 2020, the Election Infrastructure Government Coordinating Council’s executive committee issued the now-famous statement declaring that “the November 3rd election was the most secure in American history.” Chris Krebs, then running the Cybersecurity and Infrastructure Security Agency (CISA), later testified that he approved the statement and regarded it as the consensus view of the election-security community. That tidy line proved politically useful. It also sat awkwardly beside an internal intelligence record showing that multiple foreign actors had the capacity to exploit the very systems officials were publicly celebrating.
Porter, who prepared the January 2020 memo in his role overseeing cyber intelligence, says the contradiction was not an accident. “What is shocking is how uncontroversial some of these findings are to professionals—it is no secret that China and Iran compromise election equipment for a variety of intelligence purposes, nor was it controversial at the time that these systems had technical vulnerabilities,” he said. He goes further, alleging that bureaucratic and political considerations shaped what the public was allowed to know. “Every agency concurred on these findings, but because it was seen as potentially aiding the President’s reelection campaign, there was an active effort to damage him politically by refusing to share the declassified report with the public.”
Another way to put it was that the truth would have undermined faith in Joe Biden’s eventual victory. That is the heart of the whistleblower claim.
According to Porter, Trump personally ordered the information declassified because he believed election integrity demanded it. But Porter said that CIA leadership refused to release it.
“The President of the United States personally ordered this information declassified and shared with the public because he thought election integrity was so important to our country. Despite this, CIA leaders at the time refused to release the declassified report,” he said. He also alleges the resistance did not end there. “Years later, when he was reelected, CIA went so far as to claim that the report had never been declassified. Even the record of its declassification had been removed from the system,” he said. Porter describes that as an extraordinary breach of normal intelligence practice, adding, “It is important for people to recognize that this is not normal behavior by the Intelligence Community—most officers would never do something like this.”
Intelligence reports later concluded that China gained access to voter registration databases in multiple states before the election. A confidential FBI counterintelligence source also reported in summer 2020 that Beijing was attempting to interfere to aid Biden, including through a scheme involving fake U.S. driver’s licenses shipped into the country. Those reports did not become part of the public understanding in real time. Iranian hackers were not indicted until November 2021. Chinese penetration of voter data emerged publicly only after documents surfaced in March 2026. By then, the “most secure in history” line had already hardened into civic catechism.
The intelligence community’s inspector general, Christopher Fox, has opened a full investigation into whether Porter’s warnings were buried and whether he faced retaliation for pressing agencies to follow Trump’s declassification order. That review arrives alongside earlier findings from the intelligence community’s analytic ombudsman, who concluded in January 2021 that some analysts downplayed China’s role because of their disdain for Trump and reluctance to bolster his China policy.
None of this proves that foreign actors changed the 2020 outcome through hacked machines. But it tells us that senior officials knew election systems had meaningful vulnerabilities, but went out of their way to sell to the public a more politically convenient story.
Tyler Durden Tue, 04/21/2026 - 17:20Authored by Kay Rubacek via The Epoch Times,
You know this story.
Drop a frog into boiling water, and it will scramble out immediately. But place that same frog in cool water, heat it slowly, and degree by degree, it will never notice the danger until it is too late.
Most of us accept this without question.
The problem is that the story is not true.
It traces back to a German physiologist named Friedrich Goltz, who in 1869 conducted a series of experiments with a rather unusual purpose: to determine whether the soul resided in the brain or the spinal cord. He removed portions of a frog’s brain and observed what the animal could no longer do without it.
He found that a frog without its brain would sit placidly in slowly heating water and not attempt to escape. However, a normal frog, with its brain intact, would feel the rising temperature and get out.
That finding was passed around over the decades that followed, stripped of its context, and reshaped into the cautionary tale we now all repeat.
later biologists confirmed the original finding: A frog in cold water will jump out before it gets too hot.
The frog that stays in hot water is the one that can no longer think for itself.
We have been repeating that story for more than 150 years as settled truth, because it felt right, without ever stopping to ask whether it was actually true.
We accepted a false warning about the danger of not noticing gradual change, without noticing that the warning itself was false.
That should give us pause on its own.
But this month, it became more than an interesting historical footnote when a team of researchers from Carnegie Mellon University, the University of Oxford, MIT, and UCLA published a landmark study on how artificial intelligence (AI) is affecting human cognition.
The findings are fascinating, but the metaphor they chose caught many people’s attention.
They wrote of the boiling frog.
Scientists studying the effects of AI on the human mind described how the human cost of using AI could be “analogous to the ‘boiling frog’ effect, where each incremental act feels costless, until the cumulative effect becomes overwhelming to address.”
They were describing something that doesn’t arrive in a single dramatic moment, but degree by degree, use by use, in the ordinary decisions of ordinary days.
Whether knowingly or not, they used a story about an animal that only stops trying to escape once you remove its ability to think.
In their study, the researchers gave participants a series of mathematical reasoning and reading comprehension problems to solve. One group had access to an AI assistant throughout. The other worked alone. Then, without warning, the AI was removed, and everyone was tested independently on the same problems.
The AI group performed significantly worse.
That result, perhaps, is not surprising.
What is surprising is this: Every participant in the experiment had a skip button.
There was no penalty for using it, no reward for pushing through. The choice to try or to give up was entirely their own.
The AI group chose to skip at nearly double the rate.
This was not an inability to solve the problems.
It was an unwillingness to try. After just 10 minutes of having an AI system handle every moment of difficulty, something had changed in the participants’ choices. The researchers give a name to what was lost: “desirable difficulties.” It is a term from cognitive science that describes the productive struggle that, in the moment, creates a challenge and, over time, is the process by which human beings learn, grow, and develop capabilities.
The discomfort of not knowing, the resistance of a hard problem, the effort required to work through something without being handed the answer—these are not obstacles to learning. They are learning. And AI, which is designed to be maximally helpful in the immediate moment, removes them every single time.
The concerning part is not that AI makes people less capable in any permanent or measurable sense; it is that it makes people less willing to try. It erodes the willingness to push through a challenge—the very foundation on which intelligence is built and maintained. A person who never lifts anything heavy does not lose the biological capacity for strength overnight.
This is what the researchers meant when they invoked the boiling frog metaphor. They were not predicting any single catastrophic failure. They were observing an accumulation of small human surrenders.
There is a generation growing up right now who is living this on a daily basis. A recent Gallup poll found that 42 percent of Gen Z respondents believe that AI is harming their ability to think carefully and will make it harder for them to learn in the future.
These are children and young adults, still with developing human brains, forming their cognitive habits, their tolerance for difficulty, and their relationship with struggle, inside an environment that has been optimized to remove all of those things for them as efficiently as possible. The researchers warned explicitly of the risk of creating a generation that has lost the disposition to struggle productively without technological support. That is not a distant possibility. It is a trajectory in motion.
The irony at the center of all of this is that we have spent 150 years repeating a false story about how humans fail to notice gradual danger. We have repeated it uncritically, without checking, because it felt familiar and instinctively correct. Now the scientists documenting the most significant gradual cognitive shift of our time have reached for that same false story to name what they are seeing.
It’s time to rewrite the boiling frog story.
The frog with a functioning brain gets out of the water before it gets too hot.
That capacity to feel the rising temperature, to recognize what is happening, and to choose to respond is not a small thing. It is, in the context of this moment, very nearly everything.
The question worth asking is not whether we are using AI. Most people already are, and that will not change. The question is how we respond to the rising temperature.
Tyler Durden Tue, 04/21/2026 - 17:00