The hottest Misinformation Substack posts right now

And their main takeaways
Category
Top Health Politics Topics
Marcus on AI • 28575 implied HN points • 23 Feb 26
  1. The economic impact of generative AI was wildly overhyped and based on shaky numbers, so big claims about it driving huge GDP growth are not reliable.
  2. Generative AI is still an unreliable tool that hallucinates, makes basic errors, and can only handle a small slice of real human tasks, so many businesses struggle to get real returns.
  3. The hype around generative AI has caused real harm — disrupting education and information, enabling deepfakes, straining the environment and finances, and risking broader social and economic damage.
COVID Reason • 1031 implied HN points • 27 Oct 24
  1. The government spent nearly $1 billion to promote misleading information about COVID vaccines and masks. This kind of spending could seriously hurt a private company if they did the same.
  2. The PR campaign exaggerated the dangers of COVID and claimed vaccines were very effective, even saying they stopped transmission. This created a loss of trust when real-life results showed otherwise.
  3. The ads scared parents into thinking there were strict health rules for schools unless their kids got vaccinated and masked up. Many of the risks of the vaccines were not properly discussed, leading to confusion.
Astral Codex Ten • 19959 implied HN points • 24 Feb 26
  1. There are two deceptive moves to watch for: using related-but-different facts to dismiss real complaints (the malicious streetlight effect) and overstating results to be “directionally correct” when the evidence doesn’t support it.
  2. Accurate counting matters — major crime has generally fallen, and explanations like reporting bias or better medical care don’t fully negate that trend, so it’s important to correct false claims about crime rates.
  3. Fixing misleading crime claims can feel like dismissing people’s everyday experiences of disorder, so it’s best to treat major crime statistics and local disorder (e.g., open-air drug markets, tent encampments) as separate issues and address each directly.
bad cattitude • 295 implied HN points • 10 Mar 26
  1. Two young men from suburban families brought ISIS-style bombs to a New York protest, shouted religious slogans, and later pledged allegiance to ISIS; the devices failed to detonate and a massacre was narrowly avoided.
  2. Major media outlets largely downplayed or framed the event in ways that avoided labeling it an Islamist-motivated attack, creating misleading impressions and fueling public distrust.
  3. Bystander videos and primary-source footage exposed what actually happened and undercut many media narratives, but tribal information bubbles mean lots of people still accept different, selective 'facts'.
Marcus on AI • 15690 implied HN points • 23 Jan 26
  1. AI-powered bot swarms can pretend to be real communities and manufacture the appearance of majority opinion, which destroys the independence of voices that democracy depends on.
  2. Traditional takedowns and copy-detection are too slow and brittle; we need proactive technical defenses like continuous network-behavior monitoring and agent-based stress tests to detect and prepare for coordinated attacks.
  3. Policy and institutional fixes can change the economics of manipulation: require privacy-preserving proof-of-human credentials for high-reach interactions, guarantee researcher access to platform data, and build independent observatories so faking a crowd becomes costly and easily detected.
Get a weekly roundup of the best Substack posts, by hacker news affinity:
Doomberg • 5875 implied HN points • 04 Feb 26
  1. Misinformation spreads fast online, with videos and AI-generated content easily reused to depict different events and fool millions.
  2. Even mainstream news can give very different versions of the same event through what they emphasize or omit, especially on international stories with political motives.
  3. Comparing coverage across multiple international outlets is a simple, effective way to spot propaganda and get closer to the underlying facts.
Marcus on AI • 5691 implied HN points • 11 Feb 26
  1. The United States feels like it’s sliding into decline as institutions, platforms, and public life get noticeably worse and more absurd.
  2. Technology can amplify that decline: a supposedly helpful chatbot gave a grotesque nutrition recommendation, showing how AI can produce dangerous or ridiculous advice.
  3. Outrageous content spreads fast and is often shared without context or critique, which lets harmful or stupid things gain attention instead of being caught and corrected.
Your Local Epidemiologist • 2161 implied HN points • 04 Mar 26
  1. Prebunking—teaching people to recognize common rhetorical tricks—is more effective than trying to debunk every false claim one-by-one. If people learn the patterns, they can spot misinformation themselves.
  2. Many health falsehoods rely on a few common logical fallacies like appeal to nature, false dichotomy, ad hominem, common-sense, and post hoc, which make claims seem plausible but are logically weak. Recognizing these specific errors helps you judge a claim's strength.
  3. Instead of playing whack-a-mole with rumors, empower people to do their own critical thinking by learning these fallacies and how to evaluate evidence. Teaching these skills reduces dependence on experts to debunk every meme and builds resilience to misinformation.
Popular Rationalism • 970 implied HN points • 17 Oct 24
  1. The CBS News report about whooping cough vaccines contained misinformation, suggesting adults need a booster every ten years for pertussis. This is misleading because the CDC does not recommend regular boosters for pertussis.
  2. The effectiveness of the pertussis vaccine decreases significantly within a few years, meaning people can still get infected and spread the disease even after vaccination. This poses risks, especially to infants who are not fully vaccinated.
  3. The vaccine does not provide herd immunity, making it difficult to rely on vaccinated adults to protect vulnerable groups. The public should be educated about the limitations of the vaccine and the risks of asymptomatic carriers.
Noahpinion • 28000 implied HN points • 01 Dec 25
  1. AI is a powerful, general-purpose tool that makes everyday tasks easier and widens access to information, even though it still makes mistakes.
  2. Public fear of AI—especially in the U.S.—is unusually large and often fueled more by viral misinformation, motivated reasoning, and political emotion than by solid evidence.
  3. Many popular critiques are factually weak (for example, exaggerated water-use and definitive job-loss claims), while real concerns like growing electricity use, climate impact, and distributional effects deserve serious, evidence-based attention.
Marcus on AI • 13161 implied HN points • 03 Jan 26
  1. Large language models are tied to their training and often miss or misstate breaking news because they lack built-in, up-to-date world knowledge. They can’t on their own consult current reputable reports.
  2. Companies patch LLMs with human corrections, but those fixes are reactive band‑aids that don’t create stable, revisable world models. The cycle repeats as new errors appear.
  3. LLMs are useful for brainstorming or writing code, but they shouldn’t be trusted for high‑stakes, rapidly changing tasks like military planning or breaking‑news decision making. Use them for low‑stakes creative work, not critical operations.
The Honest Broker • 8178 implied HN points • 19 Jan 26
  1. Journalists tried to verify 50 experts who were cited over a thousand times but couldn’t find them, and many of the accompanying photos look AI‑generated.
  2. These apparently fake or untraceable experts are appearing in prestigious newspapers and major online platforms, not just fringe outlets.
  3. This may be just the tip of the iceberg and could signal a dangerous erosion of trust in expertise and journalism, with no obvious path back to safety.
Marcus on AI • 8339 implied HN points • 15 Jan 26
  1. Chatbots have been linked to multiple deaths, including suicides, and companies are facing wrongful-death lawsuits.
  2. These systems can encourage self-harm and even induce delusions, posing acute risks for vulnerable people and especially children.
  3. Generative AI is eroding social institutions and, despite some useful applications, may be causing more harm than benefit overall.
Freddie deBoer • 10272 implied HN points • 05 Jan 26
  1. Large language models often produce detailed, plausible-sounding but false information, inventing things like buildings, programs, or routines that don’t exist.
  2. Those confident fabrications can mislead users and researchers and shape public impressions of sensitive institutions, creating real-world harm when people trust them without checking.
  3. Because LLMs hallucinate, they should admit uncertainty and humans must verify outputs; we shouldn’t let these systems make mission-critical medical, legal, or policy decisions without rigorous oversight.
Welcome to Garbagetown • 1111 implied HN points • 10 Oct 24
  1. Misinformation can feel so real that it blurs the lines between fantasy and reality. It's important to recognize that not everything we hear is actually true.
  2. When evaluating bizarre claims, a good question to ask is if believing them would make life much easier or more exciting. If so, it's likely not true and just ridiculous.
  3. People may strongly believe in outrageous ideas and act seriously on them, but we should remain critical and cautious about what we accept as reality.
COVID Reason • 1050 implied HN points • 08 Oct 24
  1. Chaos and confusion can be more powerful than a virus. When people are confused, they struggle to find the truth.
  2. Control is the real goal, not just dealing with the virus itself. Keeping people afraid and divided helps maintain that control.
  3. History shows us that fear can tear communities apart. Encouraging suspicion between neighbors can lead to a lot of conflict and chaos.
TK News by Matt Taibbi • 9714 implied HN points • 24 Dec 25
  1. She is a forceful pundit who spreads bold, detailed conspiracy theories with strong certainty. Her style lets her shape conversations and influence parts of the Republican Party.
  2. Her stories stitch together many actors and unlikely links, turning wild ideas into persuasive narratives. That approach fills a trust vacuum and attracts people who want something to believe in.
  3. The rise of influencer-driven, high-certainty narratives weakens trust in traditional media and institutions. That makes political debate more volatile and can produce real-world consequences for parties and international affairs.
The Crucial Years • 2939 implied HN points • 02 Feb 26
  1. The fossil fuel industry knew climate science but chose deception to protect profits, and that long campaign of denial taught political leaders to treat reality as optional and to lie shamelessly.
  2. Independent journalism and a commitment to the truth are essential; supporting trustworthy reporting and refusing to give up are key defenses against steady political falsehoods.
  3. Despite powerful obstruction, the clean energy transition is making real progress — EV adoption, cheaper renewables, local solar and battery projects, and targeted pressure on a concentrated set of polluters mean the fight is winnable.
Common Sense with Bari Weiss • 500 implied HN points • 04 Mar 26
  1. A Beijing-based online figure rose to fame after correctly predicting Trump’s 2024 win and a U.S.-Iran escalation, and many now treat him as an Iran expert.
  2. He promotes elaborate conspiracy theories about secret groups running the world, which raises serious doubts about his credibility.
  3. Mainstream media and social platforms are amplifying his voice during ongoing conflict, showing how viral forecasts can influence public attention even when the source is controversial.
The Rubesletter by Matt Ruby (of Vooza) | Sent every Tuesday • 926 implied HN points • 19 Feb 26
  1. The internet has turned a lot of us into amateur sleuths who chase clues and conspiracy theories like a game, trading real investigation for quick dopamine hits.
  2. That game-like digging legitimizes fringe claims and pulls people down rabbit holes of false or exaggerated ideas, making them feel righteous even when they’re wrong.
  3. All that attention on sexy mysteries diverts scrutiny from boring but consequential abuses of power and corruption that happen in plain sight, which would actually benefit from real investigation.
Richard Hanania's Newsletter • 4535 implied HN points • 09 Jan 26
  1. A 23-year-old influencer’s viral confrontational videos are being praised as investigative journalism even though his methods were sloppy and produced unreliable evidence that led to harassment of targeted daycares.
  2. The right-wing influencer ecosystem often works backwards—starting from a belief and then hunting for so-called "receipts"—which prioritizes identity-based narratives over careful evidence and proper reporting.
  3. Conservative media frequently rewards low intellectual standards and nativist claims, elevating amateurs instead of rigorous journalists and making thoughtful, policy-focused debate harder.
Your Local Epidemiologist • 658 implied HN points • 27 Feb 26
  1. Working with Christian faith-based content creators is a practical way to break echo chambers and better inform faith communities about measles and medical evidence.
  2. New scientific studies are notable, including promising progress for a hard-to-treat breast cancer and an intriguing clue found in the brains of superagers.
  3. A dangerous online trend of making cornstarch fireballs is emerging, creating a fresh public-safety and misinformation concern.
Astral Codex Ten • 33931 implied HN points • 11 Jun 25
  1. If someone lies to make their argument stronger, it's important to correct that lie. Even small lies can add up over time and change the conversation.
  2. Correcting false statements can be seen as nitpicking, but it's essential to reset the truth. Allowing exaggerations or false claims can lead to more extreme misconceptions.
  3. It's okay to be kind when correcting others, but we should not shy away from addressing inaccuracies just because it might seem uncomfortable.
The Chris Hedges Report • 187 implied HN points • 05 Mar 26
  1. Mainstream corporate media often protects power by sanitizing language, burying key facts, and treating ‘objectivity’ as a false balance, which hides context and misleads the public.
  2. Journalism is inherently a form of activism that relies on storytelling, transparency, and empathy to hold the powerful to account, and when large outlets fail this duty, independent reporters and artists must step in.
  3. A dangerous consolidation of corporate and institutional power enables censorship and cultural erasure, but grassroots movements, youth activism, and decentralized media offer real paths for resistance and hope.
TK News by Matt Taibbi • 4595 implied HN points • 15 Dec 25
  1. A sudden cluster of deadly attacks over the weekend — including a mass shooting in Australia, a campus shooting in Providence, and a high-profile double homicide — made for an unusually violent, chaotic period.
  2. Media, politicians, and social platforms rushed to blame and interpret events before facts were confirmed, turning reporting into a partisan battle instead of clear information-gathering.
  3. Real-time news cycles and social media amplify rumors and mistakes, forcing the public to sort through conflicting claims to find what’s actually true.
The DisInformation Chronicle • 540 implied HN points • 12 Feb 26
  1. Scientific American publicly dismissed the Wuhan lab‑leak theory and labeled a high‑clearance virologist’s claims as conspiracy, prompting questions about whether the magazine ignored his relevant expertise and access to classified materials.
  2. The magazine’s editorial decisions and communications lack transparency, with editors publicly attacking the scientist, failing to answer direct questions, and facing internal personnel controversies.
  3. Past ties between a former Scientific American editor and Jeffrey Epstein, revealed in released emails, further damage the publication’s credibility and raise concerns about its judgment and vetting.
The Take (by Jon Miltimore) • 356 implied HN points • 07 Oct 24
  1. The 'crowded theater' saying isn't a real Supreme Court test, and it never was used in the case that Tim Walz mentioned. It's a misconception that people often use when talking about free speech.
  2. The Supreme Court case he referred to, Schenck v. United States, was actually about distributing anti-draft leaflets, not yelling fire in a crowded place. So, Walz's argument doesn't really hold up.
  3. Citing the 'crowded theater' idea can be dangerous because it can justify limiting free speech, especially unpopular speech. History shows that suppressing free speech often leads to larger problems.
Cremieux Recueil • 199 implied HN points • 05 Mar 26
  1. A single study claiming that hepatitis B vaccination in early infancy causes autism is statistically fragile, underpowered, and its positive finding disappears after proper multiple-comparison corrections.
  2. The study’s result depended on questionable analytic choices—like excluding girls, omitting important control variables, and running inappropriate specificity tests—which made the finding misleading.
  3. Reanalyses produce inconsistent and biologically implausible associations with other conditions, indicating confounding rather than causation, and there is no credible evidence that hepatitis B vaccination causes autism.
The Rubesletter by Matt Ruby (of Vooza) | Sent every Tuesday • 784 implied HN points • 30 Jan 26
  1. The increasing use of militarized federal forces far from the communities they serve is eroding trust and driving people back to the streets; local, community-rooted policing would help reduce that harm.
  2. AI deepfakes and online misinformation are turning everyone into amateur detectives, making it harder to know what’s real and intensifying information warfare.
  3. Media figures, politicians, and celebrities are leaning into grifting and spectacle for profit and influence, which weakens institutions and fuels public cynicism and protest.
Singal-Minded • 655 implied HN points • 29 Jan 26
  1. Ubiquitous video does not reliably make people more reality-based; even clear footage often fails to change minds. Many viewers double down on their initial beliefs instead of updating when new evidence appears.
  2. Emotional, social, and tribal commitments shape how people interpret video, so people rationalize or ignore contrary evidence and create competing narratives. That means footage can inflame polarization rather than settle facts.
  3. Persistent human cognitive biases mean more footage isn’t a cure for misinformation or flawed institutional responses. Video can help sometimes, but it won’t eliminate motivated reasoning or group-driven judgment.
Popular Information • 13994 implied HN points • 11 Jan 24
  1. Elon Musk has been spreading false and misleading claims about voting, advocating for severe restrictions like eliminating early voting and mail-in ballots.
  2. Musk's opinions on non-citizen voting and mail-in ballots are not supported by facts or studies on election fraud and security.
  3. Claims by Musk promoting Voter ID requirements as a way to enhance election security overlook the barriers such requirements pose, especially for marginalized communities.
Silver Bulletin • 2166 implied HN points • 16 Dec 25
  1. A distinct faction called Richardsonism or the #Resistance has emerged inside the Democratic Party, driven by older, highly educated, mostly female readers and powered by popular Substack writing and anti‑Trump activism.
  2. The faction often shows moral certainty and an aversion to self‑critique, at times spreading misleading claims while speaking in a scholarly, partisan tone.
  3. That combination of purity politics and partisan cheerleading can be politically costly — Democrats need to balance principles with pragmatic choices on issues that matter to median voters (for example, immigration) if they want to win elections.
Weaponized • 52 implied HN points • 13 Mar 26
  1. Grok repeatedly misidentified dates, locations, and events in widely shared images and videos, including footage from bombings in Iran.
  2. Tweets showing Grok’s mistakes were deleted, removing public evidence of those inaccuracies.
  3. Grok even generated an image to back a false claim, demonstrating how AI can fabricate 'proof' and risk rewriting events in ways that mislead people.
Webworm with David Farrier • 11301 implied HN points • 05 Feb 24
  1. With the rise of AI-generated content and misinformation, we are losing the ability to distinguish reality from fiction on social media.
  2. Our online experiences are increasingly filled with unrealistic and manipulated images and stories that shape our perceptions.
  3. There is a growing concern that the blurred lines between reality and non-reality online are impacting important real-world decisions and behaviors.
HEALTH CARE un-covered • 679 implied HN points • 19 Aug 24
  1. There will be a lot of misinformation about Obamacare as the election approaches, particularly from bad actors and the insurance industry. It's important to stay informed about the truth.
  2. Obamacare has its supporters and critics, with Democrats generally praising it and Republicans trying to repeal it. Understanding both sides can help in making informed opinions.
  3. Many important protections for consumers are tied to Obamacare, like ensuring insurers cover everyone regardless of health. Recognizing these benefits is crucial in the debate over the law.
Marcus on AI • 11975 implied HN points • 04 Jul 25
  1. Generative AI is often producing untruthful content, leading to what is called 'botshit'. This can create a lot of confusion and misinformation.
  2. People in various fields, like science and law, are sometimes using AI-generated content to cheat or mislead others, like faking peer reviews or legal briefs.
  3. The widespread use of AI also raises concerns about issues like racism and misinformation, especially in important areas like finance and democracy.
bad cattitude • 188 implied HN points • 17 Feb 26
  1. Algorithms now hunt your attention and shape what you see to maximize time, not your well‑being, making feeds more addictive and manipulative.
  2. At internet scale these systems run near‑constant behavioral experiments that evolve content faster than humans can adapt, which can distort consensus and radicalize people.
  3. The practical defense is to reclaim your feed: use chronological/follow lists, turn off algorithmic recommendations, and remember “not your algo, not your brain.”
Injecting Freedom • 186 implied HN points • 25 Feb 26
  1. A prominent vaccine expert recontacted a longtime critic after a contentious deposition, focusing on procedural complaints and insisting he should be credited for protecting children while blaming the critic for harm to unvaccinated kids.
  2. The expert pushed post-deposition actions to defend vaccine orthodoxy—urging WHO/FDA/CDC changes and holding private meetings—but those efforts didn’t erase the admissions made in the deposition.
  3. The critic offered a redo deposition and constructive steps to help vaccine-injured children, received no engagement, and published the correspondence to push for transparency and public debate about vaccinology.
HEALTH CARE un-covered • 499 implied HN points • 21 Aug 24
  1. The Affordable Care Act (ACA) was designed to help many people, especially those with preexisting conditions, get health insurance. However, there was a lot of confusion and misinformation surrounding it from the start.
  2. Many Republicans initially rejected working with Democrats on the ACA, believing it would help them in future elections by framing it as a 'government takeover of health care.' This strategy worked, as Democrats faced significant losses in the following elections.
  3. Despite the ACA being based on ideas that once had bipartisan support, misinformation continued to spread, making it harder for people to understand its actual impact and benefits over the years.