University of Pennsylvania researchers — using U.S. taxpayer dollars — are developing an artificial intelligence (AI) tool designed to “inoculate” social media users against “misinformation” about the HPV vaccine posted on social media, grant documents obtained by Children’s Health Defense (CHD) via a Freedom of Information Act (FOIA) request revealed.

The U.S. Department of Health and Human Services (HHS) is funding the $4 million “Inoculate for HPV Vaccine” randomized controlled trial running from April 2022 through March 2027. The National Cancer Institute, part of HHS, is facilitating the funding. Funding for year three was released in April.

The study is headed up by Melanie L. Kornides, associate professor of nursing at the University of Pennsylvania, whose research focuses on increasing vaccine uptake, and also on “strategies to combat misinformation.”

Kornides is joined by a team of digital health communication experts, software and program designers, social media analysts and machine learning systems experts who will help her run the “inoculation” experiment on 2,500 parents of children ages 8-12.

The team is collecting user data from YouTube, Twitter, Facebook and Instagram where people talk about HPV and using natural language processing to train an AI tool to identify “HPV misinformation,” or posts that are critical of vaccination — whether or not the information in the post is true or false.

They will then develop and test their “inoculation tool,” exposing subjects in three study arms to different types of messaging meant to make them immune to such misinformation.

A control group will get no particular messaging and two test groups will be exposed either to messaging designed to inoculate viewers against content critical of of HPV vaccines and content critical of anti-vaccine arguments.

The subjects will get “booster” doses of messaging at three and six months after their first inoculation.

If successful, the researchers wrote, this novel approach to combating health “misinformation” can be used in “wide-scale social media campaigns” addressing pandemics, childhood vaccination and other health issues.

Misinformation about the HPV vaccine, they wrote, is a major cause of “vaccine hesitancy” that circulates widely among well-meaning people, but often comes from “anti-vaccine organizations.”

Mary Holland, co-author of “The HPV Vaccine on Trial: Seeking Justice for a Generation Betrayed” told The Defender that the University of Pennsylvania research on the part of the public health industry is “a sign of weakness.”

“When you are censoring information, labeling it misinformation and smearing us, this is a sign that they’ve lost the science and are now in a verbal food fight. It’s just a sign they are going to lose,” she said.

The funding for the grant is part of a broader project within HHS to increase HPV vaccine uptake by funding research to develop messaging for providers and trusted community leaders that will convince more parents to vaccinate children against HPV.

It is one of more than 50 grants totaling more than $40 million awarded by the HHS to universities, healthcare systems and departments of public health to increase HPV vaccine uptake that CHD identified were awarded through last year.

Since then, the number of such studies has grown, with 26 new grants totaling nearly $28 million awarded in 2024 alone.

It also aligns with a new initiative by the Biden administration to fund research on AI tools to identify and censor “misinformation,” “disinformation” and “malinformation” online, according to a U.S. House of Representatives interim report released in February. Those projects are funded through the National Science Foundation.

Tweet about son’s death after HPV shot marked ‘misinformation’

Nearly half of parents have heard stories of harm from HPV vaccination, according to the grant documents, and those parents were more likely to refuse vaccination for their children.

“Fueled by misinformation from the anti-vaccine movement through social media, parents’ concerns about side effects and safety may foster reluctance to vaccinate their children,” the documents said.

In response, they are training AI to identify misinformation using methods from Kornides’ previous work identifying HPV vaccine misinformation on Twitter. In that study, Kornides and colleagues identified tweets about the HPV vaccine using keywords and coded them for their misinformation content.

“During the open coding process, we determined that almost all posts that raised concerns about HPV vaccine contained some degree of misinformation,” they found.

The most common form of “misinformation” described adverse events from vaccines. Those tweets had greater engagement than pro-HPV vaccine tweets.

The researchers broke the misinformation down into several classes.

Overall, tweets that were deemed “misinformation” were tweets that raised concerns that: the vaccine was ineffective, unsafe or caused serious adverse events, there was inadequate or falsified safety monitoring, the pharmaceutical industry was profiteering, the government conspired with interested actors to support their profiting, children were too young for the vaccine, mandates were unjust, or there were generally issues with vaccines.

There was no discussion about the veracity of the tweeted claims.

Examples of misinformation included a family tweeting that their son died after receiving the Gardasil vaccine. The tweet read: “The gardasil/hpv vaccine destroyed our lives. #gardasillkilledmyson.”

Another tweet deemed “misinformation” for raising Pharma concerns said: “Amid questions about the safety of the #HPV #vaccine #Gardasil one of the lead researchers for the #Merck drug is speaking out about its risks, benefits and aggressive marketing #vaccines #SCAM.”

The language of this tweet is taken directly from a CBS News headline.

Another “misinformation” tweet included: “Chicago teen given #HPV #vaccine from nurse practitioner in mobile van without parental consent or knowledge.” This story was also verified by CBS News.

The researchers said they also coded for “legitimate concerns,” which were not misinformation. Those were tweets that mentioned reasons cited in academic literature explaining why some people might not get vaccinated — for example, because they don’t have a provider recommendation or because a vaccine is not required for school.

‘Truth is irrelevant’

They concluded that during their study period, nearly a quarter of the HPV-related content on Twitter was “misinformation” and three-quarters of it was “supportive.”

In other words, critiques of the vaccine were deemed misinformation and comments supporting the vaccine were deemed valid.

“This just proves yet again that anything that deviates from the orthodoxy in public health is considered misinformation,” Holland said. “Truth is irrelevant. That’s shocking and disturbing. And it can kill you, period. That’s the bottom line.”

Holland said extensive research has shown that the HPV shots can be dangerous, and there is no proof from the clinical trials or the real world that they prevent cancer, as their proponents — and the grant writers for this project — claim.

“Preventing cervical cancer 40 years in the future isn’t simply unknowable,” Holland said, “It’s total speculation.” In their book, she said, Holland and her co-authors detail the failures and dangers of the HPV vaccine.

There are clear examples, she said, from the clinical trials showing that people who already had an HPV infection or had antibodies experienced inflammation and an array of inflammatory illnesses as a result.

Merck, the company that makes the Gardasil vaccine, is facing over 200 lawsuits in federal court from people who suffered a range of serious injuries after taking Gardasil, including autoimmune disorders, premature ovarian failure and cancer. There are also over 200 Gardasil injury claims pending in the Vaccine Court.

Regardless of information from Gardasil’s clinical trials, case reports of vaccine injuries, mounting evidence from across the world that demonstrates plausible associations between HPV vaccination and autoimmune conditions, information made public in ongoing lawsuits against Merck or other publicly available factual information about the HPV vaccines, in this study the researchers will use the framework developed in their Twitter research.

They will train AI to differentiate between misinformation and supportive posts across all major social media platforms and, through surveys among their research subjects, they are identifying the top 15 and then top five misinformation messages that may make parents hesitant to vaccinate their children.

They also plan to periodically review the kinds of effective “misinformation” they want to undermine.

For example, they reported in documents from the third year of the grant that the discussion on HPV vaccination has become more personal in the past few years, with more parents questioning the HPV vaccine mandates for school children.

They “see more tweets sharing personal experiences and increasing parents’ reluctance towards the HPV vaccine for their children,” the researchers said.

‘Inoculating’ parents’ attitudes against future ‘persuasive attacks’

For each platform, the researchers will develop and test supportive inoculation messages, images and videos aimed at the concerns they determine are the most likely to stop people from vaccinating their children.

The premise behind the inoculation tool is a biological metaphor suggesting that people’s attitudes can be inoculated against future “persuasive attacks,” just like an immune system can be inoculated against a disease.

Like traditional vaccines, inoculation theory works by introducing people to the potentially dangerous argument in a weakened form — not strong enough to infect them — and providing them with a well-timed, strong and persuasive argument to counteract it.

The first part of the method — the threat — is designed to make people fear that their existing position is vulnerable, motivating them to seek a protective response. It is key, they wrote, to forewarn about future misinformation before that narrative has been accepted by a person.

If the inoculation is well-timed, it can offer protection against “misinformation” for weeks to months, depending on levels of exposure and behavior, according to the documents.

For example, they wrote, they forewarn about a “threat” with messaging that says, “You may read the HPV vaccine causes paralysis.” Then they refute it with a counterargument, presented in the image below.

Their “counterarguments” or “inoculation messages” don’t present any evidence to the user. It is simply a rhetorical strategy designed to give people specific content they can use to strengthen their attitudes against change.

This approach is innovative, they wrote, because while other studies have focused on effective provider communication and educational strategies, few studies address “misinformation generated by anti-vaccine movements on social media platforms,” which they say are a major source of information for parents, especially mothers.

The content they create will instead be geared toward “encouraging vaccination and conferring resistance to antivaccination materials.”

The researchers say that if successful, their tool can be adapted to “swiftly and efficiently” address “misinformation” regarding COVID-19, childhood vaccinations and cancer treatment.

To determine whether the inoculation worked, the primary outcome measured will be medical records showing children were vaccinated. Secondary outcomes measured will be self-reported vaccination and parents’ vaccination attitudes.

However, like the other HPV messaging trials The Defender has covered, informing parents about the risks associated with the HPV vaccine is not part of the informed consent process.