169 Science claim verifications avg. score 5.1/10 67 rated true or mostly true 97 rated false or misleading
“Post-purchase dissonance significantly mediates the relationship between fear of missing out (FoMO) and order cancellation intention, with impulsive buying acting as a catalyst that transforms FoMO into post-purchase cognitive dissonance.”
The claimed mediation chain from FoMO through post-purchase dissonance to order cancellation intention is not supported as a tested model by any cited peer-reviewed study. While academic research does link FoMO to impulsive buying and impulsive buying to post-purchase cognitive dissonance, no source in the evidence base measures or tests "order cancellation intention" as an outcome. The final link relies on a practitioner blog equating returns with cancellation—a different construct. The claim presents a plausible but unverified synthesis as an established finding.
“Elephants communicate using low-frequency sounds that are often inaudible to humans and can travel long distances, as well as through body language such as ear and trunk movements.”
This claim is well-supported by peer-reviewed research and reputable science reporting across both its components. Studies published in journals and covered by outlets like Physics World, Science News, and PubMed confirm elephants produce infrasonic vocalizations (typically below 20 Hz) that travel kilometers through air and ground. The body language component — ear flapping and trunk gestures as communicative signals — is corroborated by a 2024 Smithsonian Magazine report citing peer-reviewed research in Communications Biology. The qualifier "often inaudible" accurately reflects that infrasonic components dominate long-range communication.
“Advances in technology are projected to reduce the production cost of lab-grown meat over time.”
The evidence overwhelmingly confirms that technological advances are projected to reduce lab-grown meat production costs. Peer-reviewed research documents demonstrated reductions from $437,000 to as low as $1.95/kg in optimized systems, while multiple credible sources identify specific innovations — self-sufficient cell lines, AI-driven optimization, and improved bioreactors — as drivers of continued cost declines. Current costs remain well above conventional meat prices, and significant commercialization hurdles persist, but these do not contradict the projection-framed claim.
“NASA's Artemis II mission was staged using a green screen rather than being a genuine spaceflight.”
No credible evidence supports the claim that Artemis II was staged on a green screen. The viral images and videos cited as "proof" of staging were traced to AI-generated fabrications bearing Google SynthID watermarks, while broadcast visual anomalies were explained as standard overlay and recording artifacts. Independent verification from NASA, the Canadian Space Agency, ground-station telemetry, and multiple international fact-checkers all confirm the mission was a genuine crewed lunar flyby.
“Astronauts captured the first-ever Earthset image from space.”
The Artemis II crew did photograph Earth setting behind the Moon's far side on April 6, 2026 — a genuinely notable achievement. However, the unqualified "first-ever Earthset image from space" framing overstates what authoritative sources confirm. NASA's own releases describe the photo as "reminiscent" of Apollo 8's Earthrise rather than declaring it a verified first. The "first-ever" label traces primarily to a secondary media headline, and the claim omits that uncrewed spacecraft or earlier missions may have captured similar imagery.
“Emperor penguins face a very high risk of extinction primarily due to climate change.”
The IUCN officially reclassified emperor penguins as "Endangered" on April 9, 2026 — a category defined as facing "a very high risk of extinction in the wild" — with climate-driven sea ice loss explicitly identified as the primary threat. This determination is supported by BirdLife International, the British Antarctic Survey, peer-reviewed research, and observed population declines of 10–22% since 2009. The risk is projected over decades rather than representing imminent collapse, but the claim accurately reflects the current global scientific consensus.
“Triclosan, when present in consumer products, breaks down into chloroform and dioxin upon release into the environment.”
Triclosan does degrade into both chloroform and dioxin-class compounds, but through two separate, condition-dependent pathways — not as an automatic consequence of any environmental release. Chloroform forms when triclosan contacts chlorinated water (e.g., during water treatment), while dioxin-like compounds form under UV photolysis in sunlit surface waters. The claim's core chemistry is supported by peer-reviewed research and regulatory reviews, but its blanket phrasing overstates the inevitability and omits that conversion rates are partial and other degradation products are often more prominent.
“A lake in Antarctica transformed from ocean water to freshwater over approximately 6,000 years.”
The underlying fact is real but significantly overstated. Mercer Subglacial Lake in Antarctica is indeed freshwater today and was connected to the ocean roughly 6,000 years ago. However, the claim implies a gradual, documented transformation spanning 6,000 years, when the evidence actually shows the marine connection ended around that time, after which freshening occurred over an unspecified — likely much shorter — period via glacial meltwater dilution. The "approximately 6,000 years" figure marks the age of the transition event, not the duration of a measured conversion process.
“The Earth is flat.”
Every credible source in the evidence pool — from NASA to academic institutions to science publications — directly refutes this claim. Centuries of independent empirical evidence, including horizon observations, shadow measurements, circumnavigation, and satellite imagery, conclusively demonstrate Earth is an oblate spheroid. No peer-reviewed or scientifically credible evidence supports a flat Earth model. Arguments citing ancient civilizations' beliefs or questioning observer accessibility rely on well-documented logical fallacies and do not constitute evidence for flatness.
“Fire-induced damage to plants reduces the primary energy source in an ecosystem, resulting in fewer trophic levels being supported.”
While fire can reduce plant productivity — the ecosystem's energy base — the leap to "fewer trophic levels" is not supported by the strongest available evidence. Peer-reviewed studies show that post-fire ecosystems often shift energy pathways (e.g., from detritus to algae-based food webs) or rebound through early-successional regrowth, maintaining multi-trophic structure even when individual species abundances decline. The claim conflates reduced biomass with the elimination of entire trophic levels, overstating fire's structural impact on food webs.
“Teams in the esports game Valorant that select agent compositions with balanced roles such as duelist, controller, initiator, and sentinel have a higher probability of winning compared to teams with unbalanced compositions, according to statistical analysis of professional match data as of April 2026.”
The core idea — that balanced role compositions tend to perform well in Valorant — reflects widespread community consensus, but the claim's specific attribution to "statistical analysis of professional match data" significantly overstates the evidence. The primary statistic cited (68% vs. 52% win rate) comes from a community aggregator with no disclosed methodology, and multiple data sources explicitly state they do not perform balanced-vs-unbalanced composition comparisons. The claim frames conventional wisdom as rigorous statistical proof that does not verifiably exist.
“Brain training apps improve general cognitive function beyond the specific tasks they train.”
The weight of high-quality, independent evidence contradicts this claim. Large-scale studies, systematic reviews, and meta-analyses consistently find that brain training apps produce, at best, near-transfer gains on closely related tasks — not reliable improvements in general cognitive function. A few supportive RCTs involve specific clinical populations or narrow domains and do not justify the broad, unqualified assertion. A consensus statement signed by over 70 neuroscientists has specifically warned against such generalized claims.
“Researchers deliberately fabricated a fictitious disease called Bixonimania using AI-generated preprints and found that AI systems subsequently treated it as a legitimate medical condition.”
The Bixonimania experiment is documented in an arXiv preprint and echoed by a Johns Hopkins-affiliated post, and no source contradicts its account. However, the specific claim rests on a single non-peer-reviewed preprint with no independent high-authority confirmation. The broader phenomenon — AI systems confidently elaborating on fabricated medical content — is well-established across multiple peer-reviewed studies, lending plausibility. The claim accurately reflects what was reported but should be understood as describing a preprint finding, not a peer-reviewed, independently replicated result.
“Humans perceive women as a more homogeneous group than they perceive men.”
No direct empirical evidence in the available research supports the specific claim that humans perceive women as a more homogeneous group than men. While the outgroup homogeneity effect is a well-documented general phenomenon, gender-specific research on this asymmetry has produced mixed results with no consistent finding favoring this direction. Multiple high-authority studies actually document greater actual variability among men and stronger homogenizing attitudes directed at outgroup males, contradicting the claim's premise.
“Individuals from different ideological groups systematically generate different interpretations of the same video evidence.”
Extensive peer-reviewed research confirms that ideological identity shapes how people interpret identical visual evidence, including protest footage and news imagery. The core phenomenon — partisan groups diverging in what they perceive from the same stimuli — is well-established across multiple independent studies. However, the claim slightly overstates the precision of the evidence: most direct experiments test recall, memory distortion, or still-image framing rather than controlled real-time interpretation of identical video. The word "systematically" also implies a more universal and unconditional effect than individual studies demonstrate.
“AI companion applications can provide emotional benefits equivalent to those derived from real human relationships.”
The word "equivalent" does the heavy lifting in this claim — and the evidence does not support it. While peer-reviewed studies show AI companions can produce short-term loneliness reductions comparable to brief online human chats, these effects diminish over time and do not extend to the reciprocity, vulnerability, and sustained support that characterize real human relationships. Multiple longitudinal studies find heavier AI companion use is associated with worse outcomes, including emotional dependence and deepened loneliness.
“Men have a significantly higher sex drive than women, on average.”
The best available evidence — including a large-scale 2022 meta-analysis of over 620,000 individuals — consistently finds that men report higher average sex drive than women, with a medium-to-large effect size that remains significant after adjusting for response bias. However, the observed gap is partly driven by behavioral measures like masturbation frequency, and sociocultural factors such as sexual stigma and gendered scripts may suppress women's reported desire. The claim is directionally accurate but omits these important measurement and contextual nuances.
“Women are more emotionally driven than men in decision-making contexts.”
The scientific evidence does not support the broad claim that women are more emotionally driven than men in decision-making. Peer-reviewed meta-analyses and empirical studies show that sex differences in emotional influence on decisions are small, task-specific, and inconsistent in direction — with some research finding men more susceptible to emotional spillover in financial decisions. The claim relies on conflating emotional sensitivity or neural activation with emotion-dominated choices, a logical leap that neuroscience research explicitly cautions against.
“Parallel universes exist.”
No credible scientific source supports the assertion that parallel universes are a confirmed reality. The most authoritative sources — including the Stanford Encyclopedia of Philosophy and peer-reviewed literature — consistently describe parallel universes as a feature of certain theoretical interpretations (such as the Many-Worlds Interpretation) that lack direct empirical evidence. The strongest observational candidate, bubble-collision signatures in the cosmic microwave background, has not reached statistical significance. Stating their existence as fact conflates mathematical possibility with physical confirmation.
“Social media use causally shortens human attention spans.”
Research shows a strong association between social media use and reduced attention, but the claim's assertion of causation overstates the evidence. The best longitudinal studies rule out some confounders and reverse causation, but no randomized controlled trials confirm a direct causal link. Bidirectional effects exist — pre-existing attention difficulties may also drive heavier social media use. Most studies focus on excessive or addiction-level use in children and adolescents, not typical use across all age groups. The relationship is real but not yet proven to be causal.