In the news

Recent claims, explained with evidence and context. Dates show research updates, not necessarily when events happened.

Vaccines · Research update

United States · Sources checked

Imported cases alone did not end U.S. measles elimination

The U.S. lost its measles elimination status because one chain of local transmission lasted more than a year, not simply because travelers brought in cases.

On November 17, the Pan American Health Organization announced that the United States had lost its measles elimination status. Its review found that a single chain of transmission that began in January 2025 persisted in the U.S. for more than 12 months. That finding, not simply the arrival of infected travelers, is why the status changed.

In January, a senior CDC official described losing the designation as a cost of international travel and border conditions. The Associated Press revisited that framing after the November decision. Travel can bring measles into the country, but imported cases and short outbreaks can occur without ending elimination. The distinction is whether a chain of transmission keeps going locally.

Vaccination gaps help explain why an introduction can turn into a larger outbreak. CDC reports that national MMR coverage among kindergartners fell from 95.2% in the 2019–20 school year to 92.4% in 2025–26, with lower coverage in some communities. A national average does not tell us the risk in any particular community or identify who introduced the virus.

Losing the designation does not mean every community has an outbreak, or that sustained spread is unavoidable. High vaccination coverage, surveillance and quick outbreak response help interrupt transmission. The review did not determine how much any single policy or public figure contributed to the national resurgence.

Sources and context

Link to this story
Emergency misinformation · Research update

United Kingdom · Sources checked

The November 19 blackout poster is not a National Grid warning

National Grid rejected a circulating notice predicting a major UK power outage. General emergency advice is not evidence of a scheduled blackout.

A poster predicting a major UK blackout from 4pm on November 19 is not an authentic National Grid warning. The company’s November 16 statement explicitly rejected the circulating post and said it did not expect the claimed outage. That statement is preserved in an archive of its official account.

Full Fact’s November 17 check also obtained a denial from a National Grid spokesperson. It reported that the infographic had been shared tens of thousands of times across social media. That is the fact-checker’s reported reach, not a count independently audited here. The direct statement and the fact check overlap: both report the company’s denial, rather than two separate measurements of grid conditions.

National Grid does publish ordinary preparedness advice, including keeping a torch, spare batteries and a radio available. Its guidance acknowledges that power cuts can happen. Advice for a possible emergency is not an announcement that a particular nationwide emergency has been scheduled.

This correction concerns the poster’s authenticity, not a promise that every home will keep power all day. A local fault would not authenticate a fabricated national warning. Check the relevant electricity network’s official outage information rather than treating a forwarded screenshot as an operational notice.

Sources and context

Link to this story
Anti-Muslim narratives · Research update

United States · Sources checked

An immigration law does not override the ban on religious tests for federal office

Posts invoke the 1952 McCarran-Walter Act as a way to exclude Muslims from office. Immigration powers and qualifications for office are different things.

November posts claimed President Trump wanted to use the 1952 McCarran-Walter Act to bar Muslims from public office. Archives captured on November 13 preserve versions on David J. Harris Jr.’s Facebook page and Khan Fatima’s X account. FactCheck.org documented a similar claim in 2018. These examples establish a recurring narrative, not its total audience or a confirmed government plan.

The Constitution supplies the key rule: Article VI prohibits a religious test for federal office. Congress’s immigration powers do not provide an exception to that constitutional restriction. Being Muslim is not itself a disqualification from serving in Congress.

The McCarran-Walter Act is the Immigration and Nationality Act of 1952. The State Department’s historical account describes its rules for immigration quotas, admission preferences and naturalization. FactCheck.org’s examination found no provision barring Muslims from public office. Calling it an immigration law is not an endorsement of all its provisions; its discriminatory quota system was contested at the time.

A rumor about what a president might try is separate from what a statute legally authorizes. This check addresses the claimed legal shortcut, not undisclosed intentions or the outcome of a hypothetical court case. Ordinary eligibility requirements still apply. The misleading step is presenting immigration authority as permission to exclude officeholders because of their religion.

Sources and context

Link to this story
Anti-Muslim narratives · Research update

Germany and Switzerland · Sources checked

A Berlin protest video does not show enforcement of Switzerland’s face-covering ban

Old German footage is being recaptioned as a Swiss crackdown. A real law does not make the caption true.

November 7 posts in Spanish and English described a train-station arrest video as Swiss police enforcing a face-covering ban against Muslim women. Full Fact reported on November 17 that the video had been shared thousands of times. The archived posts establish the claim’s wording and dates, not its total audience.

The footage was already circulating in March 2024 with a Berlin caption. Full Fact matched it to a pro-Palestinian protest at Berlin Central Station on March 29, using police markings and photographs. A contemporaneous photo record by Michael Kuenne/PRESSCOV/Sipa USA, distributed through Reuters Connect, independently places a protest and police detentions at that station that day. It is Sipa’s material, not a separate Reuters investigation.

Switzerland does have a federal ban on covering the face in public, effective January 1, 2025. Its Federal Office of Justice lists exceptions, including health, weather and places of worship. That real law does not turn earlier German footage into evidence of its enforcement.

This check does not determine whether the police used justified force, every participant’s actions, or how Switzerland handles other cases. The misleading step is moving an older event to another country and assigning it a different purpose. The evidence does not establish why individual users shared the caption.

Sources and context

Link to this story
Immigration and citizenship · Research update

United States · Sources checked

A SNAP eligibility estimate is not a count of undocumented recipients

The 6.3 million figure projects the effect of a proposed rule change. It does not identify 6.3 million people receiving benefits illegally.

November 13 posts recast a proposed food-assistance change as the removal of undocumented immigrants or the discovery of illegal benefit claims. Public mirrors preserve examples from @TrumpsHurricane and @teameffujoe, along with replies repeating those interpretations. These show a recurring claim, not a verified measure of its total audience.

The USDA proposal published November 10 would narrow an eligibility pathway that links SNAP to certain Temporary Assistance for Needy Families benefits. The agency estimates that 6.3 million people in 3.2 million households would lose SNAP because of income or resource limits, with the full effect projected for fiscal year 2029. That is a modeled result of changing eligibility rules, not a tally of immigration violations or proven fraud.

USDA’s eligibility guidance says undocumented noncitizens are not eligible for SNAP. Eligibility can differ within a family: New York City’s benefits agency explains that an undocumented parent can apply on behalf of eligible children without becoming eligible personally. A household’s immigration mix does not turn a child’s lawful benefit into an illegal payment to a parent.

As of November 16, this is a proposed rule, not a completed removal of 6.3 million recipients. People can disagree about the proposed limits. The estimate does not establish that current recipients broke the rules, and it also does not prove that no individual fraud occurs. The misleading step is substituting an accusation about people for a projection about policy.

Sources and context

Link to this story
Anti-Muslim narratives · Research update

United States · Sources checked

Dunkin’s holiday branding does not establish a Christmas ban

Posts claim the chain erased Christmas while promoting Ramadan and Pride. Its live website still names Christmas, and older promotions do not prove a new policy.

November posts claimed Dunkin’ had dropped Christmas while celebrating Ramadan and Pride. A Reddit post repeated the allegation that the word had disappeared from the company’s website. Snopes’ November 14 newsletter described a video making the claim as popular and placed it in the recurring “war on Christmas” narrative. These document circulation, not its total audience.

Dunkin’s October 28 announcement does use “holiday” language for its 2026 seasonal menu and Donutcracker packaging. But choosing that wording in one announcement does not establish a ban. An older promotion still publicly available on the company’s US newsroom explicitly refers to visiting on Christmas Day, contradicting the claim that the word has been erased from its website.

The dates of comparison material also matter. Dunkin’s parent company published a Pride promotion featuring a rainbow tumbler and themed doughnuts on June 7, 2021. That is evidence of a real promotion then, not proof of a nationwide 2026 campaign or of a later decision to replace Christmas.

People can prefer different seasonal wording or choose where to shop. Those preferences are separate from a factual claim that a company banned a word. This check does not establish every store’s practices or Dunkin’s internal policies. The misleading step is treating selected wording and older promotions as proof of an anti-Christmas decision; the evidence does not establish the posters’ intentions.

Sources and context

Link to this story
Anti-Muslim narratives · Research update

United States: Dearborn, Michigan · Sources checked

A Dearborn religious procession is not evidence of a Sharia-law demand

June footage of an Ashura observance was recirculated with claims about replacing American law. The original context matters.

November posts recast footage from Dearborn as Muslims demanding Sharia and rejecting American law. An archived Facebook post by David Harris Jr. linked the scene to Michigan's newly elected Muslim senator. On November 12, Spanish fact-checker Maldita documented a similar claim by commentator Javier Negre. These are dated examples of a recurring narrative, not a measurement of its total audience.

The footage predates the November election. An archived Dearborn.org Instagram post identifies it as the annual Ashura march on June 28, 2026. Maldita also matched participants to the outlet's July 1 coverage. Reposting those images with a new political caption does not change when or why the event took place.

The organizer's own program describes its June 28 March for Justice as a commemoration of Imam Hussein. Independent reporting by Michigan Public on June 29 described the annual Ashura remembrance walk and the volunteers feeding its participants. Those contemporary accounts support the religious-event context, rather than the claim that the footage records a new demand to replace American law.

This check concerns this footage, not every participant's political views or every statement ever made in Dearborn. Religious clothing, flags or rituals do not by themselves establish a legislative demand. The misleading move is attaching an unsupported political purpose and new timing to an older event; the available evidence does not establish the sharers' intentions.

Sources and context

Link to this story
Vaccines · Research update

United States: Florida study; nationwide claims · Sources checked

A vaccine comparison does not establish vaccine-caused deaths

A Florida study compared Pfizer recipients with Moderna recipients, not vaccinated people with unvaccinated people. Claims of hundreds of thousands of vaccine-caused deaths go beyond that evidence.

On November 5, Florida Surgeon General Joseph Ladapo cited a vaccine comparison to argue that thousands of Floridians died because they received Pfizer rather than Moderna. FactCheck.org's November 12 review also documented online claims extrapolating the result to hundreds of thousands of American deaths. The question is whether the study establishes that the vaccine caused those deaths.

According to CIDRAP's analysis, the Florida study compared matched Pfizer and Moderna recipients vaccinated in 2020–2021, then followed mortality for a year. It found higher mortality in the Pfizer group but included no unvaccinated comparison group. Comparing two vaccinated groups cannot by itself establish either vaccine's effect compared with receiving neither. Study coauthor Retsef Levi told The Defender that it was not designed to prove causality.

Unmeasured health differences or different protection against infection could affect the comparison. Neither explanation is established for the Florida result. An independent 2022 Veterans Affairs study found fewer infections and COVID-related hospitalizations among Moderna recipients; its difference in COVID-related deaths was statistically uncertain. That different population and outcome measure do not settle Florida's all-cause mortality finding.

The observed difference deserves investigation. It is not an established count of vaccine-caused deaths, and national extrapolation does not make it one. That does not prove either product risk-free. Our account of the Florida study relies on the reporting below; its published full text was unavailable for direct review.

Sources and context

Link to this story
Climate and energy · Research update

United Kingdom: England · Sources checked

Old coastal photos do not disprove sea-level rise

Similar-looking shorelines are not a substitute for long-term measurements that account for tides.

Two Reuters checks published on November 10 documented the same misleading comparison using different English landmarks: Whitby and St Michael’s Mount. An October 31 Facebook post about the latter used old and recent pictures to dismiss rising seas. These are examples of a recurring narrative, not a measurement of its total reach.

The photos were taken at unknown points in the tidal cycle, according to the National Oceanography Centre’s responses to Reuters. A shoreline can look similar in two pictures even when its average water level has changed. Comparing snapshots without accounting for tides cannot establish a long-term trend.

The Met Office’s State of the UK Climate 2025 assessment reports that sea level around the UK rose about 20.1 centimetres since 1901, with a likely range of 16.6 to 23.6 centimetres. The National Oceanography Centre explains that tide gauges and satellite measurements track changes over time, rather than relying on isolated views of the coast.

That national estimate is not a claim that every location changed by exactly the same amount. Local land movement and ocean conditions matter. The photos also cannot settle policy disagreements about carbon taxes or coastal protection; those choices require evidence beyond how a familiar landmark looks.

Sources and context

Link to this story
Immigration and citizenship · Research update

Ireland and Europe; international UN agreement · Sources checked

The Marrakech Compact does not set out a population-replacement plan

A widely shared video misstates the contents of the UN’s migration agreement.

The Journal reported on November 7 that a Facebook video claimed the UN’s Marrakech Compact laid out plans to displace Irish people and other Europeans. One version, posted on November 6, had been shared more than 3,000 times, according to its reporting.

The adopted compact describes a non-legally-binding framework for cooperation on migration. Paragraph 15(c) explicitly preserves states’ authority over national migration policy, subject to international law. Its 23 objectives do not include replacing European populations.

The text does support expanding pathways for regular migration. It also calls for preventing irregular migration, managing borders, combating trafficking and cooperating on safe return and readmission. Those provisions should be read together, rather than recast as an instruction to displace a population.

People can disagree about migration levels or the compact’s policies. This check concerns what the document says; it does not settle those policy choices or establish anyone’s hidden motives.

Sources and context

Link to this story
Vaccines · Research update

Spain and Russia named in the claim; worldwide airline claim · Sources checked

Travel blood-clot guidance does not establish a vaccine-based flight ban

An old airline-warning story is circulating again, but its cited health advice is about long journeys generally.

On October 30, Full Fact reported hundreds of social-media shares of claims that airlines were banning COVID-vaccinated passengers or warning them not to fly. The posts named airlines in Spain and Russia and reused a story published in June 2021, rather than a new airline announcement.

The original story cited CDC travel guidance. That guidance explains that sitting still on long journeys can raise the risk of deep vein thrombosis, including journeys by car, bus or train. It is not a vaccine-specific flight warning or evidence that an airline has banned vaccinated passengers.

Full Fact reported finding no evidence for the alleged bans. This does not mean travel-related blood clots are imaginary, settle every question about vaccine adverse effects, or establish that every traveler is fit to fly. The correction is narrower: general travel-risk advice cannot substantiate the claimed vaccine-based airline policy.

Sources and context

Link to this story
Media literacy · Research update

United States: Michigan · Sources checked

El-Sayed’s victory speech was mostly in English

A few Arabic expressions were presented as the language of the whole address. The Free Press has corrected that claim.

A November 5 Free Press post described Michigan Senator-elect Abdul El-Sayed’s victory speech as being in Arabic, pairing that description with commentary about another Muslim politician. On November 6, the outlet corrected the claim: most of the address was in English, with a few Arabic phrases. Mediaite documented both the post and correction.

The full transcript published by Rev on November 4 is overwhelmingly in English. Near the opening, El-Sayed uses “inshallah” and “alhamdulillah” and explains their meanings in English; later he addresses JD Vance as “habibi.” Those expressions do not make the whole speech an Arabic-language address.

This is a correction about what was said and how it was presented, not an endorsement of El-Sayed’s political positions. Speaking Arabic, or quoting a religious phrase, does not by itself establish support for extremism. The outlet’s error also does not establish its intent.

Sources and context

Link to this story
Climate and energy · Research update

United States · Sources checked

A state electricity-price gap does not prove what caused it

Renewable mandates can carry costs, but a state-by-state comparison cannot isolate their effect.

A November 6 FactCheck.org investigation documented Energy Secretary Chris Wright repeatedly attributing higher electricity prices to renewable-energy mandates, including at a September 23 event and in an October 28 Bloomberg interview. A price gap between groups of states is not itself a measurement of how much those mandates added to bills.

A December 2025 MIT working paper compared states over time. The association between stricter renewable standards and higher prices disappeared after accounting for persistent state differences and shared national trends. The paper is not peer reviewed and cautions that its observational approach does not establish definitive causation.

That does not mean renewable policies are free. Berkeley Lab’s June 2026 report estimated compliance costs at roughly 5% of retail electricity bills on average across states with available data, using mostly 2024 or 2025 figures, with substantial state variation. The report cautions that these estimates cover only part of the policies’ costs and benefits; they do not measure how much of the full interstate price gap the policies caused.

Rhodium Group’s April 2026 analysis identifies several interacting drivers, including natural-gas prices, grid investment, inflation and storm or wildfire costs. Whether a particular policy is worth its costs is a policy judgment. Assigning an entire interstate price difference to one policy leaves out evidence needed to establish cause.

Sources and context

Link to this story
Climate and energy · Research update

United States: Florida · Sources checked

Florida’s floods do not establish a weather-engineering plot

The weather records explain the heavy rain; they do not support claims that storms were manufactured to disrupt voting.

Claims that Florida’s early-November floods were engineered to interfere with voting are unsupported. AFP’s November 5 investigation documented a November 2 post alleging a weather attack, along with other posts making similar accusations.

The flooding was real. Florida’s November 2 emergency order described heavy rainfall from a broad low-pressure system. A National Weather Service discussion issued November 1 explained how moist Atlantic air, a rotating disturbance and repeated thunderstorms could produce dangerous rainfall over already saturated ground.

Cloud seeding is real, but it does not give people control over large weather systems. The World Meteorological Organization distinguishes limited effects under suitable cloud conditions from claims of creating rain-bearing systems or redirecting winds. It says cloud seeding has no demonstrated method for modifying floods.

The timing of a disaster is not evidence that someone caused it. This correction does not mean the floods left voting access unaffected, or that human-caused climate change cannot influence extreme weather. It addresses the specific allegation that this storm was deliberately manufactured to disrupt the election.

Sources and context

Link to this story
Elections · Research update

United States — Texas · Sources checked

The FBI says the claimed Houston ballot raid did not happen

Posts about a Houston raid used unrelated photographs. The FBI denied conducting the alleged election operation.

The FBI says it did not conduct the Houston election raid described in posts circulating on November 3, 2026. Those posts claimed agents had found more than 20,000 fraudulent ballots. AFP and PolitiFact both reported an explicit FBI denial; PolitiFact also obtained a denial from the Harris County clerk.

The photographs do not establish the alleged discovery. A photograph of FBI vehicles came from a January operation in Union City, Georgia. The Georgia Recorder published it months before the Houston claim. AFP and PolitiFact also traced a ballot photograph to San Francisco. A photograph of a real storage business in Houston is not evidence that a raid happened there.

These findings address this particular story and its images. They do not establish that every election complaint is false, or decide any other investigation. Evidence about a specific place and date is needed before treating an allegation as a confirmed discovery.

Sources and context

Link to this story
Healthcare · Research update

United States · Sources checked

A medical discount plan is not comprehensive health insurance

FTC warnings identify misleading claims about coverage and cash incentives.

In October 2026, FTC staff announced warning letters to 25 healthcare-plan marketers and lead generators. The agency described recurring misleading claims about benefits, costs and incentives, drawing on complaints and enforcement experience.

A discount program can lower some prices without providing comprehensive insurance. An advertised reward does not establish the value or extent of coverage.

The letters do not allege wrongdoing by their recipients. Twenty-five counts recipients, not proven offenders. This does not mean every discount program or incentive offer is fraudulent.

Sources and context

Link to this story
Immigration and citizenship · Research update

United States: Washington list dispute and nationwide charges · Sources checked

A flagged voter list is not a count of illegal votes

Citizenship, registration and ballots cast are different facts.

On October 26, federal prosecutors alleged that hundreds of noncitizens had registered and voted in Washington. In an October 28 response, the state’s secretary of state said a preliminary review found citizens, duplicate records and people who were inactive or not currently registered.

This describes the agencies’ positions at that time, not an independent determination of every person’s eligibility or a current litigation update. A suspected database match is not itself a verified illegal vote.

On November 2, DOJ announced charges involving 63 people nationwide for election-related offenses, including alleged unlawful voting, false citizenship claims and fraudulent registration. Its announcement includes lawful permanent residents and four people it describes as unlawfully present. Noncitizen and undocumented are not interchangeable categories.

Those charges are allegations, not convictions; defendants are presumed innocent. The number of people charged is not a count of illegal ballots or proof that every flagged record is valid. Individual allegations require case-specific evidence. Neither the state’s preliminary review nor this announcement establishes a list-wide error rate or a conclusion of zero offenses.

Sources and context

Link to this story
Media literacy · Research update

United States: North Carolina · Sources checked

AI-generated footage is not evidence a drag event happened

A North Carolina ad depicts a performer at a scene she says she never attended.

WLOS reported on October 26–27 that an AI-generated NRCC advertisement depicted drag performer Nova dancing with congressional candidate Jamie Ager. Nova told the station she had never met him or visited the property shown. WLOS documented television and YouTube distribution.

The sponsor’s October 27 statement made allegations about performances and children. Generated footage cannot establish the participation it depicts or prove abuse. Claims about real events and donations need separate evidence.

This entry concerns the depicted scene, not a finding that every underlying allegation is false. It does not determine anyone’s legal liability.

Sources and context

Link to this story
Media literacy · Research update

France · Sources checked

These fire images do not show five churches burned in one day

Old and mislocated images do not establish a coordinated attack by Muslims.

France 24’s October 30 investigation traced images used in same-day church-attack claims to separate events. Its examples included a 2019 Saint-Sulpice fire, a rubbish-bin fire outside apartments in Lyon and a fire at an encampment that did not reach the Bordeaux church.

The image matches are the investigators’ findings, not a new forensic reconstruction by this site. Real fires and damage should not be denied. The images do not substantiate the claimed sequence or collective attribution to Muslims; unknown causes remain unknown.

Sources and context

Link to this story
Climate and energy · Research update

United States · Sources checked

Solar panels and solar towers have different wildlife risks

Concentrated solar burns are a different mechanism from landing on ordinary panels.

The US Fish and Wildlife Service distinguishes photovoltaic panels from concentrating solar facilities. Birds can collide with infrastructure at either. Flying through concentrated solar flux at some concentrating facilities can also burn birds.

That distinction matters when claims about animals being burned are applied to solar panels generally. The correction is not that solar has no wildlife impacts: siting, technology and mitigation matter.

Sources and context

Link to this story
Vaccines · Research update

Indonesia · Sources checked

A vaccine trial does not establish a wildfire plot

Indonesia’s tuberculosis-vaccine trial predates the posts linking it to fires.

AFP documented October 2026 posts linking wildfires to a Gates-backed tuberculosis vaccine program. A WHO–Indonesian ministry announcement had already described vaccine development in November 2024, and the trial sponsor announced full enrollment in April 2025.

AFP reported in October that the health ministry still described a Phase 3 candidate, not an imminent rollout. The funding relationship is real, but it does not establish an organized fire–vaccine connection.

This chronology does not determine the cause of every fire or guarantee the candidate’s safety and efficacy. Smoke harms are real. This is a dated account of the trial, not a current treatment recommendation.

Sources and context

Link to this story
Vaccines · Research update

Nigeria · Sources checked

Regulatory reliance is not a blanket vaccine-testing ban

Nigeria’s reliance guideline retains regulatory authority and laboratory testing.

October 2026 posts claimed Nigeria’s US health agreement barred NAFDAC from testing drugs and vaccines. FactCheckAfrica reported the regulator’s rebuttal.

NAFDAC’s reliance guideline describes using other regulators’ work while retaining its own decisions. Its laboratory-testing section allows risk-based reductions in duplicated testing. Reliance is different from a blanket prohibition.

This does not establish that every batch is independently retested, resolve funding or data-sharing questions, or describe separately negotiated arrangements in Ghana. This entry explains the documents reviewed in October, not an individual product approval.

Sources and context

Link to this story
Media literacy · Research update

France and United Kingdom · Sources checked

An edited photo does not establish this woman’s political history

An old London protest photo was altered and paired with Paris unrest footage.

Full Fact’s October 23 comparison traced a circulated protest image to London in 2016. It found that the face of a woman filmed confronting Paris supermarket looters had been inserted into the earlier picture.

The altered image cannot establish her identity as the original protester or her past views. This does not make the separate supermarket incident fictional, nor establish the looters’ religion or immigration status. The image comparison is attributed to Full Fact.

Sources and context

Link to this story
Healthcare · Research update

United States · Sources checked

The ACA fee surplus did not begin under Biden

A flat payment is not an individualized calculation of past overcharges.

KFF’s October 23 analysis of the announced $500 ACA payments found that federal Marketplace user-fee collections exceeded obligations in fiscal years 2018–2020, before the Biden administration.

That contradicts a history attributing the accumulated surplus solely to Biden. It does not identify the exact source-year dollars used for any check or mean all surplus originated under Trump.

A fixed payment and eligibility criteria do not demonstrate a person-by-person reconciliation of past overcharges. This is a dated account of the payment explanation, not current eligibility guidance or a finding that payments are unlawful.

Sources and context

Link to this story
Climate and energy · Research update

International · Sources checked

Revised climate scenarios do not retract human-caused warming

Changing future-emissions assumptions is not a finding that past warming never happened.

The April 2026 ScenarioMIP paper revises future pathways in light of emissions trends, technology and policy. It judges the previous highest-emissions pathway implausible and also revises low-emissions assumptions.

Those are judgments about possible futures, not a retraction of observed warming or its human cause. The paper explicitly discusses climate change and policy responses.

Scenarios are conditional, with subjective plausibility judgments; they are not assigned probabilities. The revision does not mean old assumptions remain equally plausible, or that severe outcomes are impossible.

Sources and context

Link to this story
Climate and energy · Research update

Global atmospheric science; United States policy · Sources checked

The atmosphere’s warming fingerprint has not disappeared

Different stratospheric layers cannot be treated as one uniform temperature record.

An AGU Advances analysis distinguishes weaker recent cooling in the lowest stratosphere, influenced by ozone recovery, from continued cooling higher up. Its comparison finds the observed vertical temperature pattern consistent with modeled human influences.

A later Earth’s Future analysis examines tropical layers and finds that using the appropriate observations supports identification of the human warming fingerprint. The papers have overlapping authors; they are not wholly independent replications.

October reporting described an acknowledged error in a presentation chart, while a report author disputed that the chart underlay the DOE report. Those are separate questions. Neither the layer comparison nor this correction means perfect model agreement, monotonically falling temperatures in every layer, or proof of intentional fabrication.

Sources and context

Link to this story
Vaccines · Research update

New Zealand proposed market; United States developer · Sources checked

The cattle-vaccine developer describes a protein candidate

Calling the ArkeaBio project an mRNA rollout overstates the evidence.

In October 2026, AFP reported ArkeaBio’s explanation that its methane-reducing cattle vaccine candidate uses recombinant protein. The company described delivering manufactured protein to stimulate antibodies against methane-producing microbes.

The published patent includes protein and peptide vaccine examples, but also discusses nucleic-acid approaches. A broad patent is not a definitive specification for a final product. The developer’s FAQ describes an immune response, not genetic modification of cattle.

This is a dated account of a candidate in development. Company statements and funding relationships do not establish final approval, efficacy, safety or compulsory adoption. The current-formulation account is attributed to the developer through AFP, not an independent regulatory finding.

Sources and context

Link to this story
Vaccines · Research update

United States; historical October 2026 coverage discussion · Sources checked

Coverage restrictions do not establish vaccine-caused deaths

FDA approval, vaccine recommendations and insurance coverage are different decisions.

FDA approved mFLUSIVA for adults 50 and older in August 2026. TRICARE’s flu page states that it does not cover the vaccine and links coverage to CDC recommendations. That coverage rule is not itself a finding that vaccination caused deaths.

The FDA clinical review reports 102 deaths among 35,965 mRNA-vaccine recipients and 97 among 35,951 recipients of comparator flu vaccines in the pooled safety analysis. The separate category of deaths with unspecified causes was imbalanced; it must not be substituted for all-cause mortality.

FDA assessed vaccine causation as unlikely but explicitly retained uncertainty because cause-of-death information was incomplete. It described planned active surveillance. The imbalance should not be hidden, and approval does not mean zero risk. This explains the dated records, not an individual vaccination or coverage recommendation.

Sources and context

Link to this story
Media literacy · Research update

France: Marseille · Sources checked

Marseille damage did not establish blanket religious targeting

Named local authorities disputed the claim that only Jewish shops were attacked.

AFP’s October 20 reporting, carried by Le Parisien, examined a broadcast claim that Jewish-owned businesses were burned while Muslim-owned businesses were spared.

The police prefecture disputed religiously selective looting. CRIF Marseille-Provence president Bruno Benjamin described Muslim, Jewish and Christian business owners among those harmed, while allowing possible individual exceptions. The SPCJ’s reported October 18 statement also distinguished damage to Jewish-owned shops from an established religious motive.

These named accounts contradict the categorical claim. They are not an incident-by-incident audit and do not erase Jewish victims or establish that no antisemitic incident occurred anywhere. They provide no basis for inferring perpetrators’ nationality or immigration status.

Sources and context

Link to this story
Media literacy · Research update

United States: Texas; reported Wyoming registration · Sources checked

A link to an organization does not prove it sent a mailer

TENT denied authorizing Texas mailers and texts that used its links.

The Texas Tribune’s October 20–21 reporting described mailers that linked to the Transgender Education Network of Texas, or TENT, and the organization’s denial of involvement. TENT’s own October 27 update likewise denies creating, approving or distributing the material.

The Tribune reported a Wyoming LLC registration for the Texas Transgender Student Union. That is a reported registration, not evidence that the organization had genuine student membership or permission from groups referenced in the mailers.

Names, links and registration records alone do not establish who organized or financed a mailing. This historical correction concerns attribution to TENT; it does not identify a particular sponsor, declare a false flag, or claim that later authorship allegations have been resolved.

Sources and context

Link to this story
Media literacy · Research update

United States: Michigan · Sources checked

El-Sayed did disavow the Muslim Brotherhood

The reported recorded answer contradicts the claim that no disavowal occurred.

PolitiFact’s October 19 review describes the October 17 town-hall exchange: asked whether he would disavow the Muslim Brotherhood, Abdul El-Sayed answered affirmatively. Bridge Michigan also reported the disavowal and linked the exchange.

PolitiFact reports that CNN played the answer during its October 18 interview with Mike Rogers, who continued to dispute that El-Sayed had disavowed the group. Whether an answer is convincing or sufficient is an evaluative question, distinct from whether the affirmative answer occurred.

This narrow correction relies on the reported exchange. It does not settle separate allegations about historical writing, relatives or associations, infer membership, or assign a uniform legal status to all Brotherhood branches.

Sources and context

Link to this story
Healthcare · Research update

United States; Minnesota cited in the claim · Sources checked

Social transition is not surgery

Examples offered for school-secret-surgery claims describe different activities.

CNN’s October 20 review asked the White House for examples supporting claims that schools secretly send children for gender surgery. The examples supplied concerned social transition, counseling or discussion of possible treatment; they did not establish the claimed completed surgery.

Minnesota’s statutes include consent exceptions for minors living apart from parents and managing their own finances, and for emergencies where delay threatens health. Consent rules should not be reduced to a claim that every minor always needs parental consent for every service.

The material reviewed by CNN did not support the alleged school practice. This is not an exhaustive investigation of every school, an individual legal guide, or a resolution of debates about school notification and medical treatment.

Sources and context

Link to this story
Immigration and citizenship · Research update

France · Sources checked

Unrest footage does not establish an immigration takeover

A fire cannot identify the causes or participants of a national protest.

An advance CGT Airbus statement dated October 15 called for October 17 protests over wages, prices and public services. That is direct evidence of the union’s stated demands.

Footage of disorder alone cannot establish participants’ citizenship, immigration status or motives. Nor do a union’s demands explain every participant’s conduct. Crimes and responsibility require incident-specific evidence; this correction does not deny real violence.

Sources and context

Link to this story
Vaccines · Research update

Historical biomedical evidence · Sources checked

Human RSV evidence predates the chimpanzee outbreak

Archived human tissue provides evidence against the claim that RSV began in mid-1950s laboratory experiments.

Researchers studying archived human lung tissue detected RSV antigen in a case from 1931. Their 2007 pathology paper describes immunostaining with RSV-specific antibodies; the author’s full-text copy identifies the 1931 case in its figure captions.

That evidence predates the mid-1950s chimpanzee outbreak associated with the virus’s discovery. Discovery is not necessarily origin.

This was antigen detection in preserved tissue, not modern RNA sequencing. It challenges the proposed chronology without reconstructing every detail of RSV’s evolutionary origin.

Sources and context

Link to this story
Media literacy · Research update

United Kingdom: Leeds · Sources checked

The Harehills rioter allegation was withdrawn

Apologies and settlements corrected claims about Mothin Ali’s role.

ITV reported on October 16 that two commentators apologized and agreed to pay damages after alleging that councillor Mothin Ali participated in the 2024 Harehills riot. The agreed statement read in court described him assisting police, appealing for calm and putting out fires.

This was a settlement and correction, not a contested judgment establishing every fact. A person’s presence in an incident image does not establish that they caused or joined the violence.

Sources and context

Link to this story
Healthcare · Research update

United States · Sources checked

The automatic Medicare-cut claim needs an updated timeline

Congress waived the automatic reductions behind a major campaign-ad claim.

FactCheck.org’s October 16 ad review explains that deficit legislation could have triggered large automatic Medicare reductions under PAYGO, but Congress subsequently waived those cuts.

That changes the automatic-cut claim. It does not erase separate Medicaid changes, Medicare eligibility restrictions affecting some lawful immigrants or changes to cost assistance.

This is a dated explanation of one spending mechanism, not a claim that no health-coverage changes occurred or advice about current individual benefits.

Sources and context

Link to this story
Vaccines · Research update

England and United States research · Sources checked

HPV vaccination has evidence of cervical-cancer benefit

Safety signals deserve study; they do not erase observed prevention benefits.

An English population study published in 2024 found lower cervical-cancer incidence in cohorts offered HPV vaccination. This is observational evidence, with limits; the program’s results should not all be labeled Gardasil-specific.

A 2025 insurance-claims study did find associations with autonomic dysfunction and menstrual irregularities. Its dose-response findings were inconclusive. Associations warrant investigation; they do not by themselves establish infertility or that total harms outweigh prevention benefits.

This entry compares the scope of those studies, not individual medical advice or a promise of zero risk. Population, product and study design matter.

Sources and context

Link to this story
Climate and energy · Research update

United States · Sources checked

Blackouts cannot be blamed on one energy label

Official investigations identify freezing, fuel problems and preparation across technologies.

The FERC/NERC investigation of the February 2021 freeze identified freezing and fuel problems across generating technologies. A count of affected generating units is not the same as their share of lost energy.

A September 2026 review credited improved cold-weather preparedness and coordination, plus additional resources from emergency orders. Correcting selective accounts of 2021 should not erase those later contributions.

Reliability depends on equipment, fuel, preparation, transmission, demand and reserves. No fuel label guarantees an outage-free grid; this does not deny renewable intermittency.

Sources and context

Link to this story
Healthcare · Research update

United States · Sources checked

A vote on a broad care bill is not a vote on every treatment

Campaign ads can blur medication, surgery and legislative scope.

FactCheck.org’s October 12 ad analysis identified examples that inferred support for particular surgeries from votes on broader measures, including bills covering medication or other spending provisions.

An accurate account needs the actual vote, the bill’s scope and relevant amendments. Opposition to a broad bill does not by itself establish a position on every procedure it covers.

The review’s spending total described a broad advertising category, not a finding that every ad was false. Clinical and policy disputes remain separate; this entry makes no blanket claim about any candidate’s complete voting record.

Sources and context

Link to this story
Immigration and citizenship · Research update

South Africa · Sources checked

An asylum review is not automatic citizenship

South Africa’s ruling requires assessment, not automatic approval of each claim.

The Constitutional Court’s July 7, 2026 judgment held that procedural non-compliance cannot bar an asylum application without assessment of its merits. Access to assessment is not automatic refugee status or citizenship.

An October government clarification distinguished people seeking a first asylum assessment from those whose claims had finally been rejected and people who had not sought asylum.

This describes the dated ruling and clarification, not current individual legal advice. Administrative implementation can change separately, and South African law cannot be transferred to US immigration cases.

Sources and context

Link to this story
Media literacy · Research update

United States · Sources checked

An old flight-diversion video is not a new bomb threat

October captions recycled footage of a March incident.

Lead Stories traced an October 9 post about Southwest Flight 2094 to a video of a March 6, 2026 diversion. The airline told the fact-checker there had been no new incident. Contemporary reporting attributed a finding of no credible threat to the FBI.

The clip match and FBI outcome are attributed to the linked reporting, not a newly retrieved FBI finding. Reused footage does not establish a new threat or justify a conclusion about an entire religious group.

Sources and context

Link to this story
Healthcare · Research update

United States · Sources checked

Lawfully present immigrants are not undocumented immigrants

Medicaid rules distinguish legal categories and types of coverage.

CMS’s April 8, 2026 guidance on Section 71109 distinguishes categories of noncitizens and describes restrictions on federal funding, with exceptions. It does not simply describe ending ordinary federal Medicaid for undocumented immigrants.

Full benefits, emergency Medicaid and state-funded coverage have different rules. The guidance discusses changes affecting some lawfully present people as well.

This entry describes the guidance reviewed in October. It does not establish an exact number of affected Iowans or prove improper payments never happen. Check current rules and effective dates before relying on it for individual eligibility.

Sources and context

Link to this story
Sport · Research update

International database · Sources checked

A sports-results tally is not an audited count of displaced champions

Different categories and competition levels cannot be treated as one uniform measure.

The HeCheated database’s methodology includes grassroots through professional competitions and several different athlete categories. It acknowledges that in some events it cannot identify who would have won had a listed athlete not participated.

Entries, athletes, first-place finishes and displaced competitors are different quantities. A cumulative tally also cannot settle a sport-specific eligibility or fairness policy.

The exact 2,293 subtotal circulated in a September advocacy claim was neither reproduced nor disproved in this review. Database labels should not be used to infer an individual’s medical status, and this correction does not establish that all entries are invalid.

Sources and context

Link to this story
In the news