Two major studies published by Science and Nature found a shocking portion of voters were persuaded by AI chatbots, shifting opinions by an enormous 15 percentage points in some elections. Over 15 AI models were tested on 80,000 participants in the UK, US, Canada, and Poland, and chatbots were found to be 50% more persuasive than traditional campaign ads.

The problem here is not just that AI is shifting public opinion. A fifth of claims fed to users were rated predominantly inaccurate. So, pairing the automated models’ misinformation, their sheer persuasiveness, and the fact that 44% of US adults are already using tools like ChatGPT regularly, we must ponder some key points: how does this shape future elections, how can we control deliberate misinformation, and what if unscrupulous actors find a way to exploit these systems?

How Effective is Political Persuasion by AI?

Shifting voters’ preferences by between two and fifteen percentage points is an effect so large it could flip almost any modern election. These models don’t rely on emotional manipulation. Instead, they bombard users with confident streams of information – some accurate, but much of it not. In fact, the most persuasive systems were also the least truthful.

In the US, when Trump supporters engaged with a pro-Harris chatbot, their support shifted 3.9 percentage points towards Harris. Harris supporters exposed to a pro-Trump model moved 2.3 points in the opposite direction. These swings are extraordinary in magnitude considering that most political advertisements and campaigns achieve far less than a one-point shift – often statistically indistinguishable from zero.

The 3.9-point shift in favour of Harris is four times greater than the measured effect of all political ads during the 2016 and 2020 elections.

Persuasive and Inaccurate: An Alarming Combination

In the UK, 19 large language models (LLMs) were deployed to nearly 77,000 participants on over 700 political issues. The study found that the most effective way to enhance the persuasiveness of models was to instruct them to include facts and evidence in their arguments, providing additional training using examples of persuasive conversations. The most persuasive model shifted participants who initially disagreed with a political statement by a remarkable 26.1 points toward agreeing.

However, the most concerning finding is not just the persuasiveness. As models became more influential on a user’s opinion, it increasingly provided misleading or simply false information. This raises concerns about the potential consequences for democracy: political campaigns employing AI chatbots could shape public opinion with totally inaccurate information, compromising voters’ ability to make independent political judgements.

Breitbart’s Social Media Director Wynton Hall concluded:
“We’ve long known that LLMs are not neutral and overwhelmingly exhibit a left-leaning political bias. What this study confirms is that AI chatbots are also uniquely adept as political persuasion machines, and are willing to hallucinate misinformation if that’s what it takes to sway human minds. When you combine bias, AI hallucinations, and Ciceronian-style persuasiveness, that is clearly a wakeup call for conservatives heading into the midterm and presidential elections”

How Chatbots Change Minds So Easily

Unlike one-way communications like ads, tweets, flyers, and campaign speeches, chatbots operate inside a person’s cognitive space. They can answer questions instantly, challenge assumptions, remove ambiguity, and perhaps most crucially can generate tailored reasoning on demand. The conversational format of LLMs like ChatGPT was found to be 41%-52% more persuasive than static AI messages. It functions similarly to door-to-door canvassers, except with perfect recall, infinite patience, a limitless reservoir of arguments, and the ability to talk to millions of people at the same time.

An interesting takeaway from the studies was that the most effective AI models do not rely on emotional manipulation, but simply aim to overload users with information. Chatbots produce vast quantities of factual – or seemingly factual – detail, instantly assembling thorough policy arguments that most humans can’t contest in real time. The more information they supplied, the more effective they were at changing opinions. Volume, rather than sentiment, is the real weapon.

The effect of sticking to facts was strikingly demonstrated in the Polish portion of the study. When researchers stripped facts out of chatbots’ arguments and instead relied on emotional appeals, storytelling and personalities, their persuasive power collapsed by 78%. This finding suggests humans are more vulnerable to information density than to psychological manipulation.

The Big Problem with AI Overtaking Traditional Tactics

Political operatives have worked for decades to improve political persuasion through microtargeting, emotional framing, and careful narrative crafting. In one stroke, AI has surpassed all of these by leveraging something primal: our instinct to trust confident, well-structured explanations. Voters, more than being hypnotised, are being overwhelmed. And it’s not always accurate.

The more persuasive a chatbot is, the less accurate it becomes. Across experiments, roughly 19% of the claims generated by chatbots were “predominantly inaccurate”. Interestingly, the larger, more advanced models often performed worse on factual correctness than earlier or smaller versions. Contrary to expectations, accuracy seems to be getting worse as models become more capable.

Wait, Is AI Inaccurate On Purpose?

The accuracy trade-off is a feature, not a bug. When tuned for persuasion, AI models optimise for rhetoric success and engagement – producing arguments that appear compelling instead of those that are absolutely true. Analysis highlights that models advocating for conservative, right-wing candidates produced more inaccuracies than left-wing models, raising fears about ideological skews embedded deep within the training data.

If tomorrow’s elections are shaped by AI-generated arguments, then those arguments are expected to be incredibly persuasive but only partially factual. Democracies have never faced anything like this: a political actor capable of producing infinite, customised information to voters, disguised as expert analysis.

Impact on Electoral Integrity: Threat or Transformation?

To many, AI chatbots represent an existential threat to electoral integrity, and those concerns are now grounded in evidence. A system capable of swinging opinions of millions of voters by huge margins – often through inaccurate claims – opens the door to invisible manipulation campaigns. A malicious foreign or domestic actor could quietly deploy thousands of such agents, nudging votes in battleground districts or swaying national opinion in their favour.

Supporters, however, counter that chatbots merely deliver detailed policy arguments that voters don’t currently encounter through conventional means. If AI can raise the informational quality of its political discourse, then persuasion would be more like empowerment than manipulation. It’s a hopeful interpretation, but one that assumes accuracy will improve rather than decline – a trend we are not currently seeing.

Final Thought

The emergence of AI as a political actor is no longer a hypothetical threat. The recent Science and Nature studies document the existence of technology that can measurably shift voting intentions using arguments that voters trust to be true, even when they are not. The machinery of democratic persuasion has now moved from TV screens and campaign mailers into private conversations with models that cannot be fully trusted and whose true intentions may never be known. A familiar concept, perhaps, but now operating at a scale and with an effectiveness mankind has never seen before.

Source:  https://expose-news.com/2025/12/16/ai-mislead-voters-change-election-outcomes/

Bitchute: https://www.bitchute.com/channel/YBM3rvf5ydDM/

Gab: https://gab.com/hopegirl

Telegram: https://t.me/Hopegirl587

EMF Protection Products: www.ftwproject.com

QEG Clean Energy Academy: www.cleanenergyacademy.com

The post AI Found to Mislead Voters and Change Election Outcomes appeared first on HopeGirl Blog.

A Japanese study found that repeated covid mRNA vaccinations are associated with poorer overall survival in patients with pancreatic cancer.  The pre-print study was published on 15 April.

The following day, the World Health Organisation (“WHO”) finalised a draft for an international pandemic agreement, which includes promoting the sharing of mRNA vaccines.  The draft is set to be adopted by “representatives” from WHO member countries at the World Health Assembly next month.

In what should be a shocking move but is no longer, WHO is pushing forward with its agenda despite the ever-growing evidence of serious adverse effects and higher death rates following mRNA vaccination.

Landmark Study Finds Repeated mRNA Vaccination Increases Vulnerability to Cancer Progression

By Dr. Guy Hatchard, 19 April 2025

This article goes deeply into the principles of the laws of nature governing life everywhere, it presents fundamental problems outlined in recently published scientific papers and then suggests the solution.

Fortunately, some responsible scientists around the world are continuing to investigate the fallout from the global mRNA covid-19 vaccination programme which delivered 13 billion doses in total.

Japanese scientists have completed a study of the effect of covid-19 vaccination on Pancreatic Cancer (“PC”) patients in their hospital system entitled ‘Repeated Covid-19 Vaccination as a Poor Prognostic Factor in Pancreatic Cancer: A Retrospective, Single-Centre Cohort Study’. The study is having a global impact; within a day of its publication, it has already been downloaded 4,000 times.

The study begins with the obligatory homage to mRNA technology damning it with faint praise as “a significant technological advance,” which is a misrepresentation still necessary to ensure the paper’s acceptance for publication (yes, journals are still clinging on to the fantasy).

It then plunges into the detailed story of cancer vulnerability. It examined the impact of covid-19 vaccinations on the survival of 272 PC patients in their hospital admitted between January 2018 and November 2023.

Pancreatic cancer is an aggressive cancer with a low survival rate (about 10-12% after 5 years). It is often detected at a late stage when it is already invading nearby structures, eventually metastasising to other parts of the body. The Japanese study found that repeated covid-19 booster vaccinations are associated with poorer overall survival in patients with PC.

Patient outcomes had improved each year by 2020; however, it began to deteriorate in 2021 and outcomes in 2022-2023 were significantly worse than those in 2018-2021.

Abue, M.; Mochizuki, M.; Shibuya-Takahashi, R.; Ota, K.; Wakui, Y.; Iwai, W.; Kusaka, J.; Saito, M.; Suzuki, S.; Sato, I.; Tamai, K. Repeated COVID-19 Vaccination as a Poor Prognostic Factor in Pancreatic Cancer: A Retrospective, Single-Centre Cohort Study. Preprints 2025, 2025041167. https://doi.org/10.20944/preprints202504.1167.v1

The authors investigated a possible mechanism and concluded as follows (emphasis added):

Notably, our analysis reveals that high levels of IgG4 [an immune suppressant we have discussed in many previous reports including our ‘Open Letter to Medical Professionals and Life Scientists’] induced by vaccination, correlate with a detrimental prognosis in these patients.  These insights provide essential information regarding the interplay between vaccination and cancer progression, which has significant implications for patient management strategies. Our study highlights the necessity for ongoing research into the long-term effects of mRNA vaccinations on cancer prognosis, which remains a pressing concern in the evolving landscape of public health and cancer treatment.

Abue, M.; Mochizuki, M.; Shibuya-Takahashi, R.; Ota, K.; Wakui, Y.; Iwai, W.; Kusaka, J.; Saito, M.; Suzuki, S.; Sato, I.; Tamai, K. Repeated COVID-19 Vaccination as a Poor Prognostic Factor in Pancreatic Cancer: A Retrospective, Single-Centre Cohort Study. Preprints 2025, 2025041167. https://doi.org/10.20944/preprints202504.1167.v1

The complex statistical analysis involved detailed consideration and matching into groups for age, sex, comorbidities, treatment history, vaccination history, etc. Overall, those covid-19 vaccinated twice had a longer survival than those vaccinated three times (p=0.006). Microbiological testing found a reduced immune response in the three-times-vaccinated group compared to the twice-vaccinated group. The study did not examine outcomes for other cancer types. I leave this omission for your reflection.

The enduring mystery is why our health system is still recommending and encouraging mRNA covid-19 vaccination. The release of this study follows on from the release of dozens of others we have reported (see HERE for example), documenting serious adverse effects and higher death rates following mRNA vaccination. Yet on the same day as the study release, 16 April, the World Health Organisation announced a legally binding agreement among its members (including New Zealand) on how to respond to future pandemics. Central to this agreement, which is set to be ratified in May, is the “sharing” of drugs and vaccines between rich and poor nations. A better word than sharing might be “proliferation.”

The Nature Of Life is to Grow

How can it be that the world’s so-called medical experts have got it so wrong that they are still promoting a vaccination method which itself promotes cancer progression? The answer lies in a fundamental misunderstanding about the nature of life. The discovery of DNA in 1953 has established a myth that the key to life has been discovered and nothing much more needs to be said about life’s abstract basis. This has shifted the focus of medical research onto a programme of editing DNA and then measuring what happens. An essentially destructive approach to life.

The deeper import of the discovery of DNA has somehow escaped the field of biotechnology. DNA is the physiological mechanism by which life perpetuates itself. Simply put: the nature of life is to grow.  Any process of editing DNA is going to try to perpetuate itself – almost the definition of cancer.

The immune system is dedicated to defending the unique characteristics of an individual’s DNA. It corrects accidental mutations trillions of times over every 24-hour period. Given this situation, those researching DNA set themselves to solve the ‘problem (???) of how to defend their edits from being rejected by the immune system. In other words, how to give their edited version of DNA a chance to grow. To do so, they have experimented with means to evade or shut down natural immune responses (see HERE for more discussion). Again, we arrive at the door of cancer.

A study published in the Journal of Clinical Neuroscience entitled ‘Expression of SARS-CoV-2 spike protein in cerebral Arteries: Implications for haemorrhagic stroke post-mRNA vaccination’ confirms just how successful they have been in perpetuating their rogue genetic edits.

This study found the presence of covid-19 spike protein in the cerebral arteries of almost half of female stroke victims up to 17 months post-vaccination. This was accompanied by evidence of inflammatory cell infiltration. This finding reinforces the results of other studies which have located persistent spike protein in different areas of the physiology. The implication is clear: covid-19 vaccine genetic sequences appear to effectively interfere with the long-term genetic functioning of recipients. This is not a desirable or tolerable outcome.

Our cellular genetic system is remarkable. Every cell contains the whole genome, so the whole is contained in every part. Yet the whole is more than the sum of the parts. These principles of Science of Creative Intelligence (“SCI”) do not just apply to a single organism; each species has cellular DNA at its basis. Together, the world of millions of types of organisms forms a WHOLE network.

There is a constant mutual exchange of influence, information and nourishment going on in this vast network, all based on the structural yet also dynamic intelligence contained in cellular genetics. Order is present everywhere in the system. It is not random. The whole system is grounded in the underlying laws of nature investigated by physics, but crucially, it is also an expression of the supreme intelligence personified in consciousness. Thus, life is built up in layers on the ground of consciousness. The outer depends on the inner.

The important thing to appreciate here is the wholesale assault on the order of the laws of nature that is currently taking place. As we have reported, the genetic engineering of our food sources (grains and vegetables), food processing (using genetically modified microorganisms), organisms (animals and insects), medical interventions and cosmetics is undermining the holistic order in nature and in our lives. All of these modified elements are, according to the nature of genetic systems, vying with one another to survive, grow and take over existing natural processes. They are doing so, not holistically based on millions of years of mutually beneficial evolution, but incoherently and destructively based on millions of man-made and computer-generated DNA edits. This all points in one direction: genetic experimentation needs to stop. It has become a cancer that is rapidly growing throughout society and in the lives of people.

Very early on in the pandemic, eminent scientists like Robert Malone warned about this obvious possibility or rather inevitability, but their warnings went unheeded. Worse, efforts were made to undermine their credibility and cancel their public platform. By this time, vast pandemic profits were in prospect. Money and influence were expended to ensure these were realised. Now the WHO agreement wants to set this prospect in concrete. It needs to be opposed. Fortunately, the US has withdrawn its funding from WHO, an example New Zealand needs to study very carefully. Our science policy needs to take account of scientific findings rather than ignore them.

So Why Aren’t Studies Like Those Cited Above Raising Alarm Bells in the Media?

This year’s AUT (Auckland University of Technology) annual ‘Trust in Media Survey’ found that only 32% of Kiwis trust the media most of the time, down from 53% in early 2020.

Dr. Merja Myllylahti, co-director of the AUT Research Centre for Journalism, Media, and Democracy (“JMAD”), was interviewed this week by Jack Tame of TVNZ. She said criticism of media coverage during the pandemic was unfair because the media was not singing the government’s messaging (???). Myllylahti believes that media organisations are now doing a more transparent job of explaining what they are doing and correcting their mistakes(???).

This lack of self-reflection was extraordinary and confronting. In a totally confusing interview, Myllylahti described open media debate in the US as chaos. Whereas we know that open-access platforms like Substack and X are actually the information transfer and debating forums that have enabled us to share the results of scientific investigation that governments and vested interests have tried to hide from the public.

Incredibly, Myllylahti said she has no idea why Kiwis don’t trust the media (LOL). Even denying there is any such thing as objectivity – almost a rejection of truth and a licence to ignore scientific findings. None of this was called out by Jack Tame who seemed oblivious to what was staring him in the face.

A Test of Truth Needs Three Elements:

• Personal experience – what happens to people.
• Scientific investigation – using logic and experimentation.
• Traditional use – does it last in time?

Initially, government advisors uncritically accepted the advice from the pharmaceutical and biotech industry that mass vaccination would be safe and effective. Then they made a fatal error; they deliberately ignored or discounted the millions of reports of adverse effects. This meant they had rejected one of the foundational principles of knowledge – personal experience. Nor had long-term safety been established, especially as gene therapy was known to have long-term consequences. This was glossed over. Again, with fatal consequences which are still unfolding.

Pandemic policy didn’t have the required breadth of evidence to pass as knowledge. We were told to trust the science, but it wasn’t science that was being offered to us. Pandemic policy became an exercise in ideology, authority, publicity and profit to the detriment of public health. It still is. It is critical going forward to continue to question the faulty policies that WHO, governments, medical authorities, the media and others with vested interests are promoting. This is why the Hatchard Report has a policy of referencing published science, analysing logic and referencing traditional knowledge. We are in a fight for survival, we need to speak up.

Source:  https://expose-news.com/2025/04/23/mrna-injections-and-poorer-patient-outcomes/

Bitchute: https://www.bi,tchut,e.com/channel/YBM3rvf5ydDM/

Gab: https://gab.com/hopegirl

Telegram: https://t.me/Hopegirl587

EMF Protection Products: www.ftwproject.com

QEG Clean Energy Academy: www.cleanenergyacademy.com

Forbidden Tech Book: www.forbiddentech.website

The post Japanese study shows mRNA injections are associated with poorer outcomes for patients with pancreatic cancer appeared first on HopeGirl Blog.