Names and Dates
Confirm the correct spelling of names, professional titles, dates, locations, timelines, and relationships between events.
```html
Artificial intelligence can help you research, learn, compare ideas, and organize information. It can also provide inaccurate facts, invented sources, false quotations, outdated details, and misleading conclusions. Learn how to check important claims before using or publishing them.
Generative AI systems produce answers by recognizing patterns and predicting likely language. They do not automatically confirm every sentence through an authoritative source before presenting it.
An AI answer may contain accurate information, reasonable interpretation, outdated details, unsupported assumptions, and completely invented material in the same response.
The answer may also sound confident and professional. Detailed language, exact percentages, technical terminology, or a list of citations can create the appearance of authority without guaranteeing accuracy.
Verification is the process of checking whether an important claim is supported by current, relevant, and trustworthy evidence.
A creative suggestion does not require the same level of verification as medical advice, financial information, a historical quotation, or a legal claim.
Confirm the correct spelling of names, professional titles, dates, locations, timelines, and relationships between events.
Verify percentages, prices, market figures, scientific measurements, survey results, population totals, and financial data.
Confirm the exact wording, speaker, original source, publication date, and surrounding context.
Make sure the publication exists, the authors are correct, and the findings support what the AI claims.
Confirm the jurisdiction, effective date, current status, exceptions, amendments, and official legal source.
Software, AI platforms, services, pricing, policies, and capabilities can change quickly.
Health, medication, emergency, construction, electrical, mechanical, and physical safety information requires authoritative guidance.
Elections, court decisions, company leadership, conflicts, market events, regulations, and technology announcements can change rapidly.
Use this process for facts that will influence decisions, be included in published work, or affect other people.
Separate the factual statement from surrounding explanations, opinions, predictions, or recommendations.
Decide which organization, expert, document, agency, company, court, research institution, or original creator should be able to confirm the claim.
Find the primary material instead of relying only on AI summaries, social posts, copied articles, or secondhand interpretations.
Verify the title, author, publisher, institution, date, journal, document number, or URL.
Determine when the source was published and whether newer information, revised policies, later research, or recent events have changed the situation.
Do not confirm a claim from a headline, search snippet, abstract, short quotation, or isolated sentence alone.
Determine whether the original material says what the AI claimed or whether the AI exaggerated, simplified, misinterpreted, or reversed its meaning.
Use more than one trustworthy source when the issue is important, technical, disputed, or changing quickly.
Clearly distinguish confirmed facts from analysis, opinion, probability, prediction, speculation, and personal conclusion.
Save the source, title, author, date, URL, relevant passage, and the date you checked it.
A source should be evaluated according to who created it, how current it is, what evidence it provides, and whether it has a reason to mislead.
Official documents, original research, direct statements, public records, datasets, and transcripts provide evidence closest to the event or claim.
Government agencies can be authoritative for laws, regulations, statistics, public programs, safety notices, and official records.
Universities, peer-reviewed journals, professional organizations, and recognized research institutions can support technical or scientific claims.
Company documentation can confirm its own products, leadership, policies, prices, technical features, announcements, and services.
Reputable reporting can provide current context, interviews, investigation, and independent confirmation of public events.
Experts may help interpret complex evidence, but their credentials, conflicts of interest, and area of specialization should be considered.
These may offer useful insight or experience but should not automatically be treated as authoritative evidence.
Social posts may provide leads or direct statements, but screenshots, edited clips, missing context, impersonation, and rapid misinformation require caution.
Use the SOURCE method when deciding whether a webpage, document, article, video, or report deserves your trust.
Who created the information, and can the person, organization, publication, or institution be clearly identified?
Does the source provide direct documentation, data, quotations, methodology, or links to original material?
When was the information published or revised, and could the subject have changed since then?
Does the source have appropriate knowledge or authority for this particular subject?
Is the source selling something, promoting a political position, defending an organization, attracting attention, or benefiting from a particular conclusion?
Can the main claim be confirmed through additional independent and reliable evidence?
AI-generated citations may be accurate, partly correct, unrelated to the claim, or entirely fabricated.
Search for the exact title, author, publication, organization, and date.
Do not rely on the link text or AI summary. Open the source and review it directly.
Find the paragraph, table, quotation, data point, or conclusion that supports the AI statement.
Confirm who created the material and whether the publisher is reliable for that subject.
Make sure the citation is current enough for the claim being made.
A real source can still be used incorrectly if its findings, scope, or limitations are distorted.
The best verification source depends on the type of claim being checked.
Find the original speech, interview, book, article, letter, recording, transcript, or official publication.
Find the original dataset, survey, report, methodology, sample size, date, and definition of what was measured.
Locate the original research, examine the methodology and limitations, and compare later studies or expert reviews.
Use archives, museums, government records, recognized historical institutions, contemporary documents, and established scholarship.
Check the actual law, regulation, court decision, official guidance, jurisdiction, effective date, and current legal status.
Consult current official documentation, release notes, platform announcements, technical standards, and direct product testing.
Compare several reputable news sources, direct statements, public records, official announcements, and the exact date of the event.
Examine the original upload, date, location, source, metadata when available, visual inconsistencies, and independent reporting.
Imagine an AI system provides the following statement:
The statistic may be accurate, exaggerated, misquoted, taken from a limited survey, or entirely invented.
Asking several AI systems the same question can reveal disagreement, uncertainty, different assumptions, and information that needs further research.
Use the same wording and context so the answers can be compared fairly.
Look for differences in names, dates, totals, definitions, quotations, sources, and conclusions.
Differences can reveal ambiguity, outdated information, unreliable sources, or unsupported assumptions.
Ask whether the systems could be repeating the same widely circulated error.
Use the comparison to identify what must be confirmed through original or official sources.
The human researcher decides what evidence is strong enough to support the final conclusion.
AI can help locate ideas and organize research, but the author remains responsible for the accuracy of the final book, article, webpage, course, video, or report.
Record every important source, publication date, relevant passage, URL, and the claim it supports.
Do not wait until the manuscript is finished to discover that central claims lack evidence.
Use a visible placeholder such as “VERIFY SOURCE” so uncertain information is not accidentally published.
Match the exact wording to the original source and preserve the intended meaning.
Laws, prices, software features, statistics, leadership, and current events may change during the writing process.
Update webpages, future editions, digital files, and public statements when a meaningful error is discovered.
Some questions cannot be verified responsibly through online research or AI comparison alone.
Consult qualified healthcare professionals and recognized medical authorities for diagnosis, treatment, medication, emergencies, and personal health decisions.
Consult an appropriately qualified legal professional for laws, contracts, disputes, rights, obligations, deadlines, and jurisdiction-specific guidance.
Consult qualified financial, tax, accounting, or investment professionals when decisions involve significant risk or personal circumstances.
Construction, electrical, mechanical, transportation, emergency, chemical, and physical safety questions require authoritative instructions and qualified oversight.
Applications, networks, databases, user accounts, payments, and sensitive information may require review by experienced security professionals.
Research claims should be evaluated through original literature, established methodology, subject specialists, and appropriate peer review.
Search ranking does not automatically prove authority, independence, accuracy, or relevance.
Headlines and snippets may oversimplify or misrepresent the complete source.
Many webpages may repeat a claim that originated from one weak or incorrect source.
Accurate older information may no longer represent current conditions.
Searching only for evidence supporting your preferred answer can hide contradictory information and important limitations.
Multiple AI systems can reproduce the same error, especially when it is widespread online.
Financial, political, professional, or promotional interests may shape how information is presented.
A real source may exist while failing to support the AI-generated statement.
Use this checklist when you need a simple reminder of the complete verification process.
Identify the precise fact, quotation, number, date, source, or conclusion being presented.
Confirm who created the information and whether the source has appropriate authority.
Review the full document, statement, study, page, or record instead of relying on a summary.
Determine whether the information is current and whether later developments changed it.
Compare additional reliable sources and look for meaningful disagreement.
Use human judgment to decide whether the evidence is strong enough for the intended purpose.
AI systems can provide inaccurate, outdated, unsupported, or fabricated information while presenting the response with confidence.
The best source is usually the original or official evidence closest to the claim, such as a research paper, government document, court record, company announcement, transcript, dataset, or direct statement.
AI can help identify claims requiring review, but its self-evaluation should not replace independent confirmation through authoritative sources.
No. Multiple systems may repeat the same widely distributed error. Agreement can guide further research but does not prove the claim.
Verification effort should match the potential consequences. Creative brainstorming requires less checking than medical, legal, financial, academic, technical, or published factual claims.
Do not present it as confirmed fact. Remove it, qualify it clearly, continue researching, or consult a knowledgeable professional.
Verification does not end when a source is found. A person must still evaluate its relevance, context, quality, limitations, and possible impact.
Artificial intelligence can suggest questions, keywords, possible sources, competing viewpoints, and claims requiring review.
Original and authoritative sources determine whether important factual statements are supported.
The human user determines whether the evidence is sufficient and remains responsible for using, publishing, or acting on the information.
The strongest AI users do not blindly trust every answer or reject artificial intelligence entirely. They use AI to explore information, ask better questions, locate possible sources, identify uncertainty, and then verify important claims through reliable evidence.