966 published verifications avg. score 4.7/10 329 rated true or mostly true 629 rated false or misleading
“The majority of Earth's breathable oxygen is produced by marine phytoplankton rather than land plants.”
The claim overstates the scientific consensus. The most authoritative sources — including the US EPA, Woods Hole Oceanographic Institution, National Geographic, and World Ocean Review — describe phytoplankton's oxygen contribution as "about half" or "roughly equal" to land plants, not a clear majority. While some estimates range as high as 50–85%, the dominant scientific framing is approximate parity (~50/50), making the word "majority" an overstatement of what the evidence reliably supports.
“Current copyright laws are insufficient to address the ethical and legal challenges posed by generative artificial intelligence models as of March 1, 2026.”
This claim is partially true but significantly overstated. The U.S. Copyright Office concluded in 2025 that existing copyright law is "flexible enough" for AI copyrightability questions and recommended no new legislation. However, major issues—particularly whether AI training on copyrighted data constitutes fair use—remain genuinely unresolved, with landmark cases like NYT v. OpenAI still pending. The blanket claim of "insufficiency" conflates unsettled legal questions (normal in evolving areas of law) with doctrinal failure, and lumps together issues where existing law is adequate with those still being litigated.
“Sugary drinks are more harmful to dental health than solid sugary foods.”
This claim is misleading. While sugary drinks do harm teeth through both sugar-driven decay and acid erosion, the blanket assertion that they are "more harmful" than solid sugary foods is not supported by the best comparative evidence. A 2025 systematic review found solid sugary snacks carried a 3.9-fold caries risk versus only 1.56-fold for sugary beverages. Sticky and chewy sweets can cling to teeth for extended periods, creating prolonged acid attacks. The claim conflates two distinct mechanisms—caries and erosion—without evidence that drinks cause greater total dental harm.
“The Slavic peoples share a common origin.”
The claim that Slavic peoples share a common origin is well-supported by mainstream scholarship. Multiple recent ancient DNA studies (2024–2025) from leading institutions converge on a shared ancestral homeland in southern Belarus and central Ukraine. Linguistic evidence also traces all Slavic languages to Proto-Slavic. However, direct genetic evidence from the earliest Slavic core regions remains limited, and significant regional divergence occurred after expansion. The core claim is accurate, but "common origin" slightly oversimplifies a complex picture.
“Many developing nations are increasingly choosing coal power over renewable energy sources due to economic and reliability concerns.”
The claim exaggerates a real but narrow trend. While coal capacity has expanded in India and parts of Southeast Asia due to economic and reliability concerns, 87–92% of new coal capacity is concentrated in just China and India — not broadly across "many" developing nations. Moreover, coal power actually fell in both countries in 2025 for the first time in 52 years, and renewables overtook coal globally. Most developing nations are not increasingly choosing coal over renewables; the dominant trajectory is toward clean energy.
“Claude Opus 4.6 successfully built a working C compiler.”
Claude Opus 4.6 did produce a functional C compiler — a 100,000-line Rust codebase that compiles Linux 6.9, passes 99% of GCC's torture tests, and builds major projects like FFmpeg, Redis, and PostgreSQL. However, the claim omits important context: the compiler relies on GCC's assembler and linker for critical steps, independent testers found reliability issues with basic programs, it was built by 16 parallel AI agents (not one instance) with human oversight, and it cost ~$20,000 in API usage. It works, but with significant caveats.
“Bed rotting improves mood and reduces stress more effectively than maintaining normal daily productivity during periods of burnout.”
This claim is false. No controlled studies compare bed rotting to normal daily productivity for burnout recovery. The Sleep Foundation explicitly states the trend "hasn't yet been directly studied by researchers," making the "more effectively" assertion unsupported. Clinical sources warn that prolonged inactivity can worsen mood and deepen a depression-inactivity cycle. While brief, intentional rest may offer some short-term relief, evidence-based burnout recovery guidelines favor active strategies like structured rest and light exercise — not extended passive inactivity.
“Honeybees can be trained to detect landmines.”
The claim is well-supported. Multiple peer-reviewed studies and government-funded research programs (including Sandia National Laboratories and DARPA) have demonstrated that honeybees can be classically conditioned to respond to TNT and other explosive odors associated with landmines, with successful field tests confirming detection capability. The research spans two decades and includes both active conditioning and passive biomonitoring approaches. However, this training enables area-level surveying of minefields rather than precise pinpointing of individual buried mines.
“Consuming fresh fruit does not typically result in excessive sugar intake for most people.”
This claim is well-supported. The WHO, ADA, CDC, NHS, and Harvard all consistently affirm that fresh fruit sugars — packaged with fiber and nutrients — do not constitute excessive sugar intake for most people at typical consumption levels. WHO guidelines explicitly exclude whole fruit from free-sugar reduction targets, citing no evidence of adverse effects. Minor caveats apply: people with diabetes or insulin resistance may need to monitor fruit intake, and very high-sugar fruits in large portions can add up. But the claim's "typically" and "most people" qualifiers accurately reflect the scientific consensus.
“Sea level is not uniform across different locations on Earth.”
Sea level is indeed not uniform across Earth's locations. Authoritative sources from NASA, NOAA, and oceanographic institutions confirm that ocean surface height varies globally due to currents, winds, gravity fields, and other physical factors.
“Kristi Noem stated that the Bible was written in English.”
This claim is false. Kristi Noem never said the Bible was written in English. The quote originated from a clearly labeled satirical Facebook page called "Mrs. Putin," which included deliberate misspellings of Noem's name and a fake "Fox Mews" logo. Multiple independent fact-checkers — including Snopes and Media Bias Fact Check — confirmed no credible record of Noem making this statement exists. The viral spread of the quote does not make it real.
“Deglobalization trends pose a significant threat to long-term economic growth in Western nations.”
The claim captures a real concern — trade fragmentation and rising tariffs do create growth headwinds for Western economies, as the IMF and OECD have documented. However, it overstates the evidence. The IMF's January 2026 outlook projects tariff drag waning and Western growth remaining resilient. Multiple institutions (ECB, J.P. Morgan, World Bank) find globalization is reconfiguring, not reversing. Academic evidence on deglobalization's growth effects is mixed. The claim treats a plausible risk as an established significant long-term threat, which the evidence does not yet support.
“The prevalence of mental health issues among young adults in Western countries has significantly increased due to social media use.”
The claim overstates the evidence. While WHO surveillance data and meta-analyses confirm correlations between heavy or "problematic" social media use and worse mental health indicators, the effect sizes are small and multiple longitudinal studies find no significant causal link. The word "due to" implies proven causation that the research does not support. Rising mental health concerns among young people likely involve multiple factors — including pandemic disruption, economic stress, and increased diagnostic awareness — not social media alone.
“Generative AI models consistently produce factual inaccuracies in their outputs.”
Generative AI models do produce factual inaccuracies, and this is a well-documented, persistent challenge confirmed by peer-reviewed research and major benchmarks. However, the word "consistently" overstates the problem. Error rates vary enormously — from below 1% on grounded summarization tasks to over 30% on open-domain reasoning — depending on the task, domain, model, and whether retrieval tools are used. Hallucination rates are also declining over time. The claim describes a real issue but frames it in a misleadingly uniform way.
“A tomato is botanically and/or culinarily classified as a vegetable.”
The claim is largely accurate on its culinary prong: multiple authoritative sources (PubChem/NIH, Britannica, U.S. legal precedent) confirm tomatoes are considered vegetables for culinary, nutritional, and legal/customs purposes. However, the botanical prong is clearly false — tomatoes are botanically classified as fruits (specifically berries), not vegetables. Because the claim uses "and/or," only one prong needs to hold, and the culinary classification is well-established. The inclusion of "botanically" is misleading but does not invalidate the overall statement.
“Sound can have a negative decibel level.”
Sound can indeed have negative decibel levels. The decibel scale uses a logarithmic ratio formula, so any sound intensity below the chosen reference point mathematically produces a negative dB value. This is confirmed by multiple academic physics sources and occurs in both digital audio systems and theoretical acoustic measurements.
“Teaching students according to their preferred learning styles (visual, auditory, or kinesthetic) improves educational outcomes.”
This claim is not supported by scientific evidence. Multiple high-quality meta-analyses and reviews — including a 2024 PMC meta-analysis and publications from the APA, AFT, and leading cognitive science journals — consistently find no convincing evidence that matching instruction to students' preferred learning styles (visual, auditory, kinesthetic) improves educational outcomes. The "meshing hypothesis" is widely classified as a neuromyth by cognitive scientists. Academic performance is better explained by factors like background knowledge, motivation, and study strategies.
“The "five-second rule" — the belief that food dropped on the floor is safe to eat if picked up within five seconds — prevents the transfer of harmful bacteria to food.”
The five-second rule does not prevent harmful bacteria from transferring to dropped food. Peer-reviewed research, including a comprehensive 2016 Rutgers study, shows bacteria can transfer to food in less than one second upon contact. While longer contact times may increase contamination, there is no safe window. Factors like moisture, surface type, and contamination level often matter more than time. The claim is not supported by scientific evidence.
“Wearing sunscreen with SPF is recommended for skincare and skin protection even when indoors.”
The claim is partially true but misleadingly broad. Reputable medical sources like MD Anderson and Keck Medicine of USC do recommend sunscreen indoors — but specifically when you spend prolonged time near windows, since UVA rays can penetrate glass. However, Cancer Council Australia and other authorities say indoor sunscreen is "typically" unnecessary because overall UV exposure indoors is low. The blanket phrasing "even when indoors" overstates what is actually a conditional recommendation tied to window proximity, skin conditions, and exposure duration.
“Live sports broadcasts cannot be convincingly deepfaked using current technology as of March 1, 2026.”
This claim is false. As of March 2026, real-time deepfake systems can already generate convincing manipulations of sports footage at broadcast frame rates (40–50 FPS) on both datacenter and consumer hardware. While limitations remain with extreme camera angles and multi-person occlusions, these are partial constraints — not fundamental barriers. Convincing deepfakes of live sports segments, interviews, and selective broadcast shots are demonstrably achievable today, making the blanket assertion that they "cannot" be done inaccurate.