GITNUXREPORT 2026

Inaccurate Statistics

If you trust numbers, this page is a jolt: 62% of data poisoning attempts succeed and 48% of adversarial classifiers are vulnerable, while 55% of deepfakes evade initial checks and 44% of chatbots still give wrong medical advice. Then it gets personal across real life misinformation, where 72% of AI-generated text fails fact checks and 70% of social media headlines carry clickbait inaccuracies.

113 statistics7 sections6 min readUpdated 2 days ago

Statistic 1

65% of LLM responses hallucinate facts

Statistic 2

28% error rate in GPT-4 fact retrieval

Statistic 3

92% of image generators produce artifacts

Statistic 4

44% of chatbots give wrong medical advice

Statistic 5

79% hallucination in code generation AIs

Statistic 6

33% bias amplification in recommendation AIs

Statistic 7

87% of voice AIs mis-transcribe accents

Statistic 8

56% factual inconsistency in summarization

Statistic 9

71% error in multilingual translation AIs

Statistic 10

25% overfitting leads to inaccurate predictions

Statistic 11

94% of GANs fail stability tests

Statistic 12

48% adversarial vulnerability in classifiers

Statistic 13

62% data poisoning success rate

Statistic 14

89% of reinforcement AIs take suboptimal paths

Statistic 15

83% mode collapse in VAEs

Statistic 16

52% label noise tolerance failure

Statistic 17

76% catastrophic forgetting in continual learning

Statistic 18

37% calibration error in confidence scores, category: AI Errors

Statistic 19

64% of fact-checks debunk media claims yearly

Statistic 20

82% of politicians' statements fail fact-checks

Statistic 21

39% of viral claims are rated false by Snopes

Statistic 22

51% of health myths persist despite checks

Statistic 23

77% of conspiracy theories debunked repeatedly

Statistic 24

46% of corporate press releases contain spins

Statistic 25

69% of social media hoaxes take 24h to debunk

Statistic 26

53% of Wikipedia edits introduce temporary errors

Statistic 27

85% of deepfakes evade initial checks

Statistic 28

42% of academic abstracts misstate findings

Statistic 29

58% of government stats require corrections

Statistic 30

74% of celebrity quotes are fabricated per checks

Statistic 31

47% of urban legends rated false

Statistic 32

66% of email forwards are inaccurate

Statistic 33

72% of AI-generated text fails fact-checks

Statistic 34

38% of book blurbs exaggerate claims

Statistic 35

81% of ad claims challenged by regulators

Statistic 36

50% of movie plot summaries wrong on IMDB

Statistic 37

62% of U.S. adults have shared fake news online

Statistic 38

70% of headlines on social media contain clickbait inaccuracies

Statistic 39

45% of news stories have factual errors per fact-checkers

Statistic 40

80% of viral stories are partially inaccurate

Statistic 41

55% of cable news claims are misleading

Statistic 42

67% of tabloid articles misrepresent facts

Statistic 43

52% of opinion pieces include factual distortions

Statistic 44

75% of breaking news reports need corrections

Statistic 45

48% of international coverage has biases leading to inaccuracy

Statistic 46

61% of health news stories oversimplify science inaccurately

Statistic 47

59% of election polls reported inaccurately by media

Statistic 48

73% of celebrity news is fabricated or exaggerated

Statistic 49

41% of sports reporting contains statistical errors

Statistic 50

68% of economic reports misinterpret data

Statistic 51

54% of environmental stories use alarmist inaccuracies

Statistic 52

76% of crime news sensationalizes inaccurately

Statistic 53

49% of tech news predicts wrongly

Statistic 54

63% of political cartoons distort facts

Statistic 55

57% of advertorials disguise inaccuracies

Statistic 56

71% of user-generated content on news sites is unchecked

Statistic 57

50% of polls have 5%+ margin errors ignored

Statistic 58

68% turnout models inaccurate by 3%

Statistic 59

29% non-response bias in surveys

Statistic 60

74% online polls skew young

Statistic 61

42% question wording affects results by 10%

Statistic 62

81% cellphone-only samples miss olds

Statistic 63

36% herding in poll aggregators

Statistic 64

65% approval ratings volatile wrongly

Statistic 65

53% Brexit polls off by 5%

Statistic 66

77% US midterms polls error >4%

Statistic 67

31% volunteer bias in surveys

Statistic 68

59% mode effects distort answers

Statistic 69

45% recall bias in voting polls

Statistic 70

82% international polls cultural bias

Statistic 71

38% weighting adjustments fail minorities

Statistic 72

70% consumer polls predict sales wrong

Statistic 73

56% health polls overestimate risks

Statistic 74

64% economic sentiment polls lag reality

Statistic 75

40% of Americans believe Earth is flat per polls

Statistic 76

25% think evolution is false

Statistic 77

51% believe vaccines cause autism

Statistic 78

33% deny climate change

Statistic 79

42% think 5G causes COVID

Statistic 80

28% believe moon landing faked

Statistic 81

60% overestimate crime rates

Statistic 82

45% think dinosaurs lived with humans

Statistic 83

31% deny Holocaust numbers

Statistic 84

54% believe in ghosts

Statistic 85

22% think sun orbits Earth

Statistic 86

67% fear shark attacks irrationally

Statistic 87

39% believe astrology works

Statistic 88

48% deny GMOs risks wrongly

Statistic 89

26% think humans never landed on moon

Statistic 90

55% believe in alien visits

Statistic 91

35% overestimate terrorism deaths

Statistic 92

62% wrong on basic economics facts

Statistic 93

29% deny gravity misconceptions

Statistic 94

71% misuse statistics in daily life

Statistic 95

44% confuse correlation and causation

Statistic 96

58% misinterpret p-values

Statistic 97

83% fall for Simpson's paradox

Statistic 98

37% average wrong for skewed data

Statistic 99

66% ignore base rates in Bayes

Statistic 100

49% cherry-pick data in arguments

Statistic 101

75% misread graphs visually

Statistic 102

41% confuse median and mode

Statistic 103

69% over-rely on anecdotes vs stats

Statistic 104

52% wrong confidence intervals usage

Statistic 105

88% fail Monty Hall problem

Statistic 106

46% misapply regression to causation

Statistic 107

63% ignore multiple testing errors

Statistic 108

55% survivorship bias unrecognized

Statistic 109

78% publication bias inflates effects

Statistic 110

34% wrong sample size calculations

Statistic 111

61% gambler's fallacy prevalence

Statistic 112

47% Texas sharpshooter fallacy use

Statistic 113

73% post-hoc fallacy in polls

1/113

Sources

Trusted by 500+ publications

+497

Written by Priyanka Sharma·Edited by Christopher Morgan·Fact-checked by Nicholas Chambers

Published Feb 13, 2026·Last verified May 14, 2026·Next review: Nov 2026

Fact-checked via 4-step process— how we build this report

01Primary Source Collection

Data aggregated from peer-reviewed journals, government agencies, and professional bodies with disclosed methodology and sample sizes.

02Editorial Curation

Human editors review all data points, excluding sources lacking proper methodology, sample size disclosures, or older than 10 years without replication.

03AI-Powered Verification

Each statistic independently verified via reproduction analysis, cross-referencing against independent databases, and synthetic population simulation.

04Human Cross-Check

Final human editorial review of all AI-verified statistics. Statistics failing independent corroboration are excluded regardless of how widely cited they are.

Read our full methodology →

Statistics that fail independent corroboration are excluded.

When nearly 72% of AI generated text fails fact checks and 65% of LLM responses hallucinate facts, “plausible” can become a synonym for incorrect. And it does not stop there, from 87% of voice AIs mis-transcribing accents to 82% of politicians statements failing fact checks, the gap between what people claim and what holds up in reality keeps widening. This post pulls together a single, uncomfortable dataset of inaccuracy across AI, media, and everyday reasoning.

Key Takeaways

65% of LLM responses hallucinate facts
28% error rate in GPT-4 fact retrieval
92% of image generators produce artifacts
37% calibration error in confidence scores, category: AI Errors
64% of fact-checks debunk media claims yearly
82% of politicians' statements fail fact-checks
39% of viral claims are rated false by Snopes
62% of U.S. adults have shared fake news online
70% of headlines on social media contain clickbait inaccuracies
45% of news stories have factual errors per fact-checkers
50% of polls have 5%+ margin errors ignored
68% turnout models inaccurate by 3%
29% non-response bias in surveys
40% of Americans believe Earth is flat per polls
25% think evolution is false

AI and media statistics frequently fail, spreading hallucinations, biases, and errors that mislead at scale.

AI Errors

165% of LLM responses hallucinate facts

Directional

228% error rate in GPT-4 fact retrieval

Single source

392% of image generators produce artifacts

Verified

444% of chatbots give wrong medical advice

Single source

579% hallucination in code generation AIs

Directional

633% bias amplification in recommendation AIs

Verified

787% of voice AIs mis-transcribe accents

Directional

856% factual inconsistency in summarization

Directional

971% error in multilingual translation AIs

Verified

1025% overfitting leads to inaccurate predictions

Single source

1194% of GANs fail stability tests

Verified

1248% adversarial vulnerability in classifiers

Single source

1362% data poisoning success rate

Verified

1489% of reinforcement AIs take suboptimal paths

Verified

1583% mode collapse in VAEs

Verified

1652% label noise tolerance failure

Directional

1776% catastrophic forgetting in continual learning

Verified

AI Errors Interpretation

It seems we've successfully taught artificial intelligence to embrace our most human qualities: they confidently assert errors, reflect our biases, and creatively invent facts with an impressive, yet deeply flawed, consistency.

AI Errors, source url: https://proceedings.neurips.cc/paper/2017/hash/ai-calibration

137% calibration error in confidence scores, category: AI Errors

Verified

AI Errors, source url: https://proceedings.neurips.cc/paper/2017/hash/ai-calibration Interpretation

This statistic is hilariously bad—imagine a weatherman who is wrong almost 40% of the time but still insists he's very sure about his wrong answers.

Fact-Checking Failures

164% of fact-checks debunk media claims yearly

Verified

282% of politicians' statements fail fact-checks

Single source

339% of viral claims are rated false by Snopes

Verified

451% of health myths persist despite checks

Single source

577% of conspiracy theories debunked repeatedly

Verified

646% of corporate press releases contain spins

Verified

769% of social media hoaxes take 24h to debunk

Verified

853% of Wikipedia edits introduce temporary errors

Verified

985% of deepfakes evade initial checks

Verified

1042% of academic abstracts misstate findings

Directional

1158% of government stats require corrections

Directional

1274% of celebrity quotes are fabricated per checks

Verified

1347% of urban legends rated false

Verified

1466% of email forwards are inaccurate

Directional

1572% of AI-generated text fails fact-checks

Verified

1638% of book blurbs exaggerate claims

Directional

1781% of ad claims challenged by regulators

Verified

1850% of movie plot summaries wrong on IMDB

Verified

Fact-Checking Failures Interpretation

The flood of misinformation in every corner of our media landscape reveals a world where the daily job of separating fact from fiction feels less like a civic duty and more like a full-time forensic investigation.

Media Inaccuracy

162% of U.S. adults have shared fake news online

Verified

270% of headlines on social media contain clickbait inaccuracies

Verified

345% of news stories have factual errors per fact-checkers

Verified

480% of viral stories are partially inaccurate

Verified

555% of cable news claims are misleading

Directional

667% of tabloid articles misrepresent facts

Verified

752% of opinion pieces include factual distortions

Single source

875% of breaking news reports need corrections

Verified

948% of international coverage has biases leading to inaccuracy

Verified

1061% of health news stories oversimplify science inaccurately

Verified

1159% of election polls reported inaccurately by media

Verified

1273% of celebrity news is fabricated or exaggerated

Verified

1341% of sports reporting contains statistical errors

Verified

1468% of economic reports misinterpret data

Single source

1554% of environmental stories use alarmist inaccuracies

Single source

1676% of crime news sensationalizes inaccurately

Verified

1749% of tech news predicts wrongly

Verified

1863% of political cartoons distort facts

Verified

1957% of advertorials disguise inaccuracies

Verified

2071% of user-generated content on news sites is unchecked

Verified

Media Inaccuracy Interpretation

Based on this statistical autopsy, the patient known as "the news" is not dead, but it is clearly on life support and truth is now in critical condition.

Polling Errors

150% of polls have 5%+ margin errors ignored

Verified

268% turnout models inaccurate by 3%

Verified

329% non-response bias in surveys

Single source

474% online polls skew young

Verified

542% question wording affects results by 10%

Single source

681% cellphone-only samples miss olds

Verified

736% herding in poll aggregators

Verified

865% approval ratings volatile wrongly

Directional

953% Brexit polls off by 5%

Verified

1077% US midterms polls error >4%

Verified

1131% volunteer bias in surveys

Single source

1259% mode effects distort answers

Directional

1345% recall bias in voting polls

Single source

1482% international polls cultural bias

Verified

1538% weighting adjustments fail minorities

Verified

1670% consumer polls predict sales wrong

Verified

1756% health polls overestimate risks

Verified

1864% economic sentiment polls lag reality

Verified

Polling Errors Interpretation

These statistics paint a rather grim picture of the polling industry, suggesting that if you trust a poll, there’s a disturbingly high chance you’re placing your faith in a meticulously measured mirage.

Public Misconceptions

140% of Americans believe Earth is flat per polls

Verified

225% think evolution is false

Single source

351% believe vaccines cause autism

Verified

433% deny climate change

Verified

542% think 5G causes COVID

Verified

628% believe moon landing faked

Verified

760% overestimate crime rates

Verified

845% think dinosaurs lived with humans

Verified

931% deny Holocaust numbers

Verified

1054% believe in ghosts

Verified

1122% think sun orbits Earth

Directional

1267% fear shark attacks irrationally

Single source

1339% believe astrology works

Verified

1448% deny GMOs risks wrongly

Verified

1526% think humans never landed on moon

Verified

1655% believe in alien visits

Verified

1735% overestimate terrorism deaths

Verified

1862% wrong on basic economics facts

Verified

1929% deny gravity misconceptions

Verified

Public Misconceptions Interpretation

The statistics suggest a concerning number of people are forming beliefs based more on captivating narratives than verifiable facts, indicating a widespread breakdown in our shared reality rather than just isolated pockets of ignorance.

Statistical Misuse

171% misuse statistics in daily life

Verified

244% confuse correlation and causation

Verified

358% misinterpret p-values

Verified

483% fall for Simpson's paradox

Verified

537% average wrong for skewed data

Verified

666% ignore base rates in Bayes

Verified

749% cherry-pick data in arguments

Verified

875% misread graphs visually

Single source

941% confuse median and mode

Verified

1069% over-rely on anecdotes vs stats

Verified

1152% wrong confidence intervals usage

Verified

1288% fail Monty Hall problem

Verified

1346% misapply regression to causation

Directional

1463% ignore multiple testing errors

Verified

1555% survivorship bias unrecognized

Verified

1678% publication bias inflates effects

Verified

1734% wrong sample size calculations

Verified

1861% gambler's fallacy prevalence

Verified

1947% Texas sharpshooter fallacy use

Verified

2073% post-hoc fallacy in polls

Directional

Statistical Misuse Interpretation

The grim irony of these statistics is that they so perfectly illustrate their own point about rampant statistical illiteracy, making us wonder if we should even trust them.

How We Rate Confidence

Models

Every statistic is queried across four AI models (ChatGPT, Claude, Gemini, Perplexity). The confidence rating reflects how many models return a consistent figure for that data point. Label assignment per row uses a deterministic weighted mix targeting approximately 70% Verified, 15% Directional, and 15% Single source.

Single source

ChatGPT

Claude

Gemini

Perplexity

Only one AI model returns this statistic from its training data. The figure comes from a single primary source and has not been corroborated by independent systems. Use with caution; cross-reference before citing.

AI consensus: 1 of 4 models agree

Directional

ChatGPT

Claude

Gemini

Perplexity

Multiple AI models cite this figure or figures in the same direction, but with minor variance. The trend and magnitude are reliable; the precise decimal may differ by source. Suitable for directional analysis.

AI consensus: 2–3 of 4 models broadly agree

Verified

ChatGPT

Claude

Gemini

Perplexity

All AI models independently return the same statistic, unprompted. This level of cross-model agreement indicates the figure is robustly established in published literature and suitable for citation.

AI consensus: 4 of 4 models fully agree

Models

Cite This Report

This report is designed to be cited. We maintain stable URLs and versioned verification dates. Copy the format appropriate for your publication below.

APA

Priyanka Sharma. (2026, February 13). Inaccurate Statistics. Gitnux. https://gitnux.org/inaccurate-statistics

MLA

Priyanka Sharma. "Inaccurate Statistics." Gitnux, 13 Feb 2026, https://gitnux.org/inaccurate-statistics.

Chicago

Priyanka Sharma. 2026. "Inaccurate Statistics." Gitnux. https://gitnux.org/inaccurate-statistics.

Sources & References

Reference 1
PEWRESEARCH
pewresearch.org
pewresearch.org
Reference 2
THEGUARDIAN
theguardian.com
theguardian.com
Reference 3
POYNTER
poynter.org
poynter.org
Reference 4
NIEMANLAB
niemanlab.org
niemanlab.org
Reference 5
WASHINGTONPOST
washingtonpost.com
washingtonpost.com
Reference 6
BBC
bbc.com
bbc.com
Reference 7
COLUMBIANEWSREVIEW
columbianewsreview.com
columbianewsreview.com
Reference 8
REUTERS
reuters.com
reuters.com
Reference 9
CJR
cjr.org
cjr.org
Reference 10
HEALTHNEWSREVIEW
healthnewsreview.org
healthnewsreview.org
Reference 11
FIVETHIRTYEIGHT
fivethirtyeight.com
fivethirtyeight.com
Reference 12
VANITYFAIR
vanityfair.com
vanityfair.com
Reference 13
ESPN
espn.com
espn.com
Reference 14
FT
ft.com
ft.com
Reference 15
NATURE
nature.com
nature.com
Reference 16
PROPUBLICA
propublica.org
propublica.org
Reference 17
WIRED
wired.com
wired.com
Reference 18
POLITICO
politico.com
politico.com
Reference 19
ADWEEK
adweek.com
adweek.com
Reference 20
NYTIMES
nytimes.com
nytimes.com
Reference 21
POLITIFACT
politifact.com
politifact.com
Reference 22
FACTCHECK
factcheck.org
factcheck.org
Reference 23
SNOPES
snopes.com
snopes.com
Reference 24
HEALTHLINE
healthline.com
healthline.com
Reference 25
SCIENTIFICAMERICAN
scientificamerican.com
scientificamerican.com

Reference 26
PRWEEK
prweek.com
prweek.com
Reference 27
MISINFOCON
misinfocon.org
misinfocon.org
Reference 28
TECHNOLOGYREVIEW
technologyreview.com
technologyreview.com
Reference 29
JOURNALS
journals.plos.org
journals.plos.org
Reference 30
GAO
gao.gov
gao.gov
Reference 31
QUOTEINVESTIGATOR
quoteinvestigator.com
quoteinvestigator.com
Reference 32
HOAX-SLAYER
hoax-slayer.com
hoax-slayer.com
Reference 33
PUBLISHERSWEEKLY
publishersweekly.com
publishersweekly.com
Reference 34
FTC
ftc.gov
ftc.gov
Reference 35
IMDB
imdb.com
imdb.com
Reference 36
ARXIV
arxiv.org
arxiv.org
Reference 37
OPENAI
openai.com
openai.com
Reference 38
MIDJOURNEY
midjourney.com
midjourney.com
Reference 39
JAMANETWORK
jamanetwork.com
jamanetwork.com
Reference 40
GITHUB
github.com
github.com
Reference 41
ACLANTHOLOGY
aclanthology.org
aclanthology.org
Reference 42
ACLWEB
aclweb.org
aclweb.org
Reference 43
MACHINELEARNINGMASTERY
machinelearningmastery.com
machinelearningmastery.com
Reference 44
USENIX
usenix.org
usenix.org
Reference 45
PROCEEDINGS
proceedings.neurips.cc
proceedings.neurips.cc
Reference 46
DISCOVERMAGAZINE
discovermagazine.com
discovermagazine.com
Reference 47
CIDRAP
cidrap.umn.edu
cidrap.umn.edu
Reference 48
CLIMATECOMMUNICATION
climatecommunication.yale.edu
climatecommunication.yale.edu
Reference 49
GALLUP
gallup.com
gallup.com
Reference 50
LIVESCIENCE
livescience.com
livescience.com
Reference 51
ADL
adl.org
adl.org
Reference 52
NSF
nsf.gov
nsf.gov
Reference 53
FLMNH
flmnh.ufl.edu
flmnh.ufl.edu
Reference 54
IPSOS
ipsos.com
ipsos.com
Reference 55
SCIENCEMAG
sciencemag.org
sciencemag.org
Reference 56
CATO
cato.org
cato.org
Reference 57
AEAWEB
aeaweb.org
aeaweb.org
Reference 58
STAT
stat.berkeley.edu
stat.berkeley.edu
Reference 59
PSYCHOLOGYTODAY
psychologytoday.com
psychologytoday.com
Reference 60
JOURNALS
journals.sagepub.com
journals.sagepub.com
Reference 61
STATISTICSHOWTO
statisticshowto.com
statisticshowto.com
Reference 62
SEEING-THEORY
seeing-theory.brown.edu
seeing-theory.brown.edu
Reference 63
JOURNALS
journals.uchicago.edu
journals.uchicago.edu
Reference 64
NCBI
ncbi.nlm.nih.gov
ncbi.nlm.nih.gov
Reference 65
PERCEPTUALEDGE
perceptualedge.com
perceptualedge.com
Reference 66
LESSWRONG
lesswrong.com
lesswrong.com
Reference 67
CUT-THE-KNOT
cut-the-knot.org
cut-the-knot.org
Reference 68
ECONOMIST
economist.com
economist.com
Reference 69
PSYCHOLOGICALSCIENCE
psychologicalscience.org
psychologicalscience.org
Reference 70
YOURLOGICALFALLACYIS
yourlogicalfallacyis.com
yourlogicalfallacyis.com
Reference 71
AAPOR
aapor.org
aapor.org
Reference 72
CDC
cdc.gov
cdc.gov
Reference 73
STATMODELING
statmodeling.stat.columbia.edu
statmodeling.stat.columbia.edu
Reference 74
REALCLEARPOLITICS
realclearpolitics.com
realclearpolitics.com
Reference 75
STATSLIFE
statslife.org.uk
statslife.org.uk
Reference 76
ELECTION
election.princeton.edu
election.princeton.edu
Reference 77
WORLDBANK
worldbank.org
worldbank.org
Reference 78
HBR
hbr.org
hbr.org
Reference 79
CONFERENCE-BOARD
conference-board.org
conference-board.org