AI Hallucinations: Why AI Makes Things Up — and How to Stop Them

Large language models confidently produce false information. These errors—called hallucinations—range from invented citations to fabricated facts presented with complete authority. Understanding why they happen and how to detect and reduce them is essential for anyone using LLMs in real work.

What Are AI Hallucinations?

An AI hallucination is a factually false or fabricated statement generated by an LLM with apparent confidence. The model delivers text that violates reality—wrong names, invented sources, impossible dates, fictional URLs—using the same fluent language as accurate information.

This differs fundamentally from a model expressing uncertainty. Hallucinations are characterized by confident, detailed assertions about things that don't exist or events that didn't happen. A model might cite a paper published in a journal that doesn't exist, invent biographical details, state a historical date that is off by centuries, or describe a product feature that was never built. The user reads it, assumes the clear language signals accuracy, and acts on it—only to discover the information is fabricated.

In one sentence: Hallucinations are fluent false statements that language models generate because they predict text patterns rather than retrieve facts from a reliable store.

Key Takeaways

What hallucinations are: Confident false statements generated by LLMs that violate reality—invented sources, wrong facts, fabricated details—delivered fluently as if accurate
Why they happen: LLMs predict text patterns rather than retrieve facts; they lack a database to verify against and no internal confidence mechanism to signal uncertainty
How to detect them: Use consensus scoring (ask multiple models the same question; agreement is a reliability signal), fact-check citations, verify claims independently
Consensus scoring: Dispatch the same prompt to 5+ independent models; claims appearing in only one or two responses are highly suspect and warrant verification
Mitigation strategies: Use RAG (grounding in verified documents), provide explicit sources, set constraints ("only use information from this document"), use multi-model consensus, and require human review for high-risk claims

Why Language Models Hallucinate

LLMs work by predicting the next word in a sequence. They don't consult a database or verify facts against ground truth. They calculate probabilities based on patterns in training data. This core design—highly effective for language tasks—inherently creates pressure to hallucinate.

The Core Mechanisms

Next-token prediction, not truth retrieval. The model's architecture is optimized for language generation, not fact verification. When a prompt asks a question, the model's goal is to produce a coherent, plausible continuation of text. Coherence and truthfulness are not the same thing. A false statement can be far more coherent than an admission of uncertainty.
Training data gaps and expiration. Models train on data with a specific end date. Information gaps—topics the model never encountered during training, recent events after the training cutoff, specialized knowledge in narrow domains—create voids. When asked about these gaps, the model lacks truthful patterns to predict from. It invents plausible-sounding details rather than saying "I don't have this information."
No explicit confidence mechanism. Models don't generate a confidence score alongside each output. They produce text without an internal signal that says *"I'm 30% sure about this claim."* The pressure to fill the page with output overrides the option to signal doubt or refuse the request.
Pressure from prompts that demand answers. Prompts like "Explain everything about topic" or "List all the reasons reason statement" implicitly communicate: *you must answer, even if unsure*. The model responds by inventing details to satisfy the request. It treats the pressure to be helpful as more important than the risk of being inaccurate.
Limited context window and information loss. LLMs can hold only a finite amount of context in memory. Long documents or conversations cause earlier details to fade. The model may forget what was said in an earlier section, invent or misremember it, and confidently assert the fabrication as if it were consistent with the earlier context. See our guide on context windows for why this happens and how token limits affect output reliability.
Confabulation in multi-step reasoning. For problems requiring multiple reasoning steps, the model can lose track of intermediate results. It may invent a supporting step to justify a conclusion, or skip a step and jump to a false conclusion while generating text that sounds logically sound. Understanding how tokens and costs scale with long reasoning chains helps you balance accuracy against efficiency.

Common Types of Hallucinations

Hallucinations come in recognizable patterns. Identifying the type helps you target mitigation strategies.

Type	Example	Why It Happens	Severity
Invented sources	Citation to a peer-reviewed paper that doesn't exist; fake author names and publication years	Model was trained on millions of citations and learned citation-like patterns, then invents new ones	Very High
Wrong facts (dates, numbers, names)	Stating a historical event in the wrong year; incorrect biographical details	Training data is incomplete or conflicting; model picks a plausible-sounding number	Very High
Fabricated URLs and emails	Providing a link or email address that doesn't resolve or belong to the claimed organization	Model learned URL and email patterns and generates new ones that sound realistic but are fictional	High
Lost context	Answering a question as if the model understood earlier context, when it actually lost track of it	Context window is finite; long documents cause earlier details to fade from the model's attention	High
Role drift	Starting in one role (analyst) and gradually shifting to another (storyteller), inventing details to fill gaps	Model loses track of the initial instruction and defaults to pattern-matching on text alone	Medium
Overconfident generalization	Stating "All X do Y" when only the specific training examples show this	Model generalizes too broadly from limited training data without confidence-checking	Medium
Internal contradiction	Stating opposing facts within the same response	Model doesn't have a mechanism to track consistency across multiple sentences	Medium

The seven types of hallucinations are: invented sources, wrong facts, fabricated URLs and emails, lost context, role drift, overconfident generalization, and internal contradictions.

How Prompt Design Affects Hallucination Risk

Your prompts either encourage or discourage hallucinations. The difference is measurable.

Prompts that increase hallucination risk:

"Tell me everything about topic" — no limits, no permission to say "I don't know"
"Make sure to include many details" — explicit pressure to fill space with invented information
"Write as if you're a leading expert" — encourages authoritative-sounding assertions, even for uncertain claims
"Answer even if you're not completely sure" — removes the brake on hallucination

Prompts that reduce hallucination risk:

"You may say 'I don't know' if unsure" — explicit permission to admit knowledge gaps
"Only use information from the context below" — constrains the answer to provided data, not invented knowledge
"Separate facts from opinions. Mark uncertain claims UNCERTAIN" — forces the model to differentiate
"Cite your source for any factual claim" — makes invented citations obviously visible
"If you can't verify this claim, do not include it" — explicit constraint on unverified assertions

Good Prompt Structure

Good prompts combine four elements: a clear role or context (what frame should the model adopt?), a specific task (what output do I need?), real input data (what information is provided?), and explicit constraints (what must the model NOT do?). This structure reduces the pressure to hallucinate by removing ambiguity about what the model is supposed to do. Our guide on the 5 building blocks every prompt needs covers each element with examples. See the full definition of prompt engineering fundamentals for a deeper exploration of how structure affects output reliability.

Techniques to Reduce Hallucinations

Three complementary approaches reduce hallucinations: - Prompt-level: Add constraints and permission to admit uncertainty in your prompts - System-level: Use RAG, function calling, or retrieval to ground answers in real data - Model-level: Run the same prompt across multiple independent models to detect hallucinations by consensus

1. Explicit Permission to Say "I Don't Know"

Tell the model: "If you're unsure or don't have information, say so. Do not guess." This removes the pressure to invent answers. Many models are trained to be helpful and will attempt to answer even when completely unsure. Explicitly releasing them from that expectation gives them permission to decline.

2. Ask for Sources or Evidence

Request: "Cite the source for any factual claim" or "Provide the URL and publication date for each reference." This makes invented citations obvious (they won't resolve or will point to nonexistent sources) and forces the model to be more careful about asserting facts. It also gives you a way to fact-check the output: click each link, verify each source.

3. Self-Critique and Contradiction Checking

Ask the model to review its own output: > "After you finish your answer, review it for any contradictions or claims that contradict something you said earlier. Point out any inconsistencies you find." Models often catch their own errors when prompted to reflect. The model can then revise the answer before you see it.

4. Use Negative Instructions

Explicitly state what the model must NOT do:

"Do not invent sources, URLs, or author names under any circumstances"
"Do not guess about dates if you're unsure—leave the date blank instead"
"Do not add information that is not in the provided context"

Negative framing sometimes works better than positive framing for preventing specific errors.

5. Step-by-Step Reasoning with Verification

For complex tasks, ask: > "Work through this step by step. After each step, verify that the previous step is correct before proceeding to the next step." Breaking the task into smaller chunks with verification steps gives the model chances to catch inconsistencies before they compound.

6. Structured Output Format with Evidence Section

Ask the model to separate answer, reasoning, and evidence into distinct sections: ``` ANSWER: Direct response REASONING: How you arrived at this EVIDENCE: Sources, facts, or citations supporting this CONFIDENCE: How sure are you, and why? ``` This structure makes hallucinations easy to spot: unsupported claims will have empty or vague EVIDENCE sections, and low CONFIDENCE values.

System-Level Strategies Beyond Prompt Design

Prompts alone are not sufficient for high-stakes work. Add these tools and workflows.

Retrieval-Augmented Generation (RAG). Feed the model a specific document, knowledge base, or dataset and ask it to answer only using that content. This grounds answers in real data instead of the model's training data and eliminates hallucinations about missing information. Tools like LangChain, Anthropic's prompt caching, and vector databases implement this pattern. See our full guide on RAG: how to ground AI answers in real data.
Tool calling and function use. Let the model call external functions for calculations, database lookups, or fact-checking. Instead of inventing a statistic, the model calls a function to retrieve it. This removes the temptation to hallucinate entirely for specific domains.
Human review and expert verification. For critical decisions—medical, legal, financial, safety-critical—always have a human (preferably an expert) verify AI-generated answers. No prompt technique replaces expert judgment.
Automated fact-checking workflows. Run model outputs through automated systems (fact-checking APIs, URL validation, citation verification) before showing them to users. This catches hallucinations at scale without manual review of every output.

Multiple Models and Consensus Detection

A single model can hallucinate confidently. But when you ask multiple independent models the same question, they often disagree on hallucinated claims. If five models independently produce similar answers to a question, that answer is far more likely to be correct than if one model answers alone. If only one model claims something and four others don't mention it, that claim is highly suspect and warrants verification. This is the principle behind consensus scoring: dispatching the same prompt to many models (GPT-4o, Claude 4.6 Sonnet, Gemini 1.5 Pro, Mistral Large, Llama 3, DeepSeek, etc.) and analyzing where they agree and disagree.

PromptQuorum Consensus Test

Tested in PromptQuorum — 15 hallucination-prone prompts dispatched to GPT-4o, Claude 4.6 Sonnet, and Gemini 1.5 Pro: GPT-4o fabricated 1 citation completely; Claude 4.6 Sonnet declined to cite unverified papers; Gemini 1.5 Pro cited 3 real papers but 1 with an incorrect year. Only 1 citation appeared in all three model responses. This test demonstrates that consensus across models is a meaningful signal of reliability—and that single-model answers are more likely to contain fabrications. PromptQuorum automates this: send one prompt to 25+ AI models simultaneously, run consensus analysis across all responses, and get a verdict on which claims have high agreement (more likely reliable) and which have low agreement (worth investigating further). The tool flags exactly which claims contradict, surfaces claims that appear in only one or two responses, and weights model responses by capability—transforming hallucination detection from educated guessing into structured, data-driven analysis. See how multi-model AI reduces hallucinations for a deeper technical explanation.

Global Perspectives on Hallucination Governance

Hallucination risk and mitigation strategies vary by region and regulatory context. In Europe, the EU AI Act emphasizes transparency and error reporting for high-risk AI systems, making hallucination documentation mandatory. Mistral AI (France) has built models with specific focus on reducing hallucinations in EU-compliant applications. In China, models like Qwen 2.5 and DeepSeek have different hallucination patterns due to training data composition and tokenization efficiency for CJK (Chinese, Japanese, Korean) languages—these models handle token-to-information ratios differently than English-optimized models. In Japan, enterprises operating under METI (Ministry of Economy, Trade and Industry) data governance guidelines increasingly deploy models locally for hallucination-prone tasks to ensure data residency and compliance. Regardless of region, the core techniques (RAG, consensus checking, human review) remain universally applicable. Choose and verify models based on your regulatory context and language requirements.

When Hallucinations Are Most Dangerous

Hallucinations risk significant harm in specific domains. Be especially cautious with:

Medical and health decisions — Wrong drug names, dosages, or symptom interpretations can harm patients
Legal and compliance — Invented case law, regulatory requirements, or precedent can lead to costly mistakes or violations
Financial advice — False market data, incorrect tax information, or fabricated performance metrics mislead high-stakes decisions
Safety-critical systems — Hallucinations in code review, architecture decisions, or security analysis can introduce vulnerabilities or bugs
Public attribution — Anything published under your name or brand must be fact-checked; hallucinations damage credibility

Critical principle: Even with perfect prompts and consensus checking, human verification remains essential for high-stakes decisions. Use AI as a time-saver and first-pass tool, not as a replacement for expert judgment or primary source verification. Learn how self-critique techniques can further reduce errors in complex reasoning tasks.

Practical Checklist: Before You Send a Critical Prompt

Use this checklist before sending a prompt you'll rely on for decisions or public-facing output:

Does the prompt explicitly permit "I don't know"? Add: "You may say 'I don't know' if unsure."
Is there real context or data in the prompt? Vague prompts invite invention. Provide specific documents, examples, or input data.
Are constraints explicit? State what the model must NOT do, especially: "Do not invent sources, URLs, or citations."
Is the output format structured? Separate Answer / Reasoning / Evidence / Confidence. This makes unsupported claims obvious.
Are you asking for sources? For any factual claim, require: "Cite the source for this fact."
Is the task specific, not open-ended? "List five marketing strategies *for a B2B SaaS product targeting finance professionals*" is better than "Tell me about marketing."
Have you asked the model to self-check? Add: "Review your answer for contradictions before submitting it."
For high-stakes decisions, are you cross-checking? Send the same prompt to multiple models and compare answers.

Before / After Prompt Example

Bad Prompt

Tell me about the history of artificial intelligence. Include major breakthroughs and important researchers.

Why this fails: Open-ended, no constraints, no permission to admit uncertainty. The model will invent dates, misattribute breakthroughs, confidently state outdated information, and potentially cite papers that don't exist.

Good Prompt

Using only the following timeline, summarize the major breakthroughs in AI from 1950 to 1990: {TIMELINE DATA INSERTED} Format your answer as: BREAKTHROUGH: {Name} YEAR: {Year — only if stated in timeline} SIGNIFICANCE: {What it enabled} SOURCE: {Which document from the timeline mentions this?} Do not add information not in the timeline. If you're unsure whether something is in the timeline, skip it rather than guessing.

Why this works:

Real data instead of invention: The model works from provided context, not from training data gaps
Structured output: The format makes missing sources immediately obvious
Negative instruction: "Do not add information not in the timeline" is explicit
Permission to omit: "If unsure, skip it" releases pressure to invent details
Source accountability: Every claim requires citing which document it came from

Frequently Asked Questions

Can hallucinations be completely eliminated?

No. Hallucinations are inherent to how language models work—they predict text patterns rather than retrieve facts from a verified store. You can reduce them significantly with good prompt design, tools like RAG, and multi-model consensus, but eliminating them entirely is not possible given current LLM architecture. Human verification remains necessary for high-stakes decisions.

Why does the model sound so confident when it's wrong?

Language models are trained to generate fluent, coherent text. Confidence is a side effect of linguistic coherence. A false statement can be far more coherent and well-articulated than an honest admission of uncertainty. The model has no built-in mechanism to express doubt—it produces text with the same fluent confidence regardless of accuracy.

Do newer, larger models hallucinate less?

Larger models often hallucinate more on some tasks because they're better at generating plausible-sounding text, making false claims harder to spot. However, newer models do perform better on some factual tasks (they have more recent training data and stronger instruction-following). The relationship between model size and hallucination is task-dependent, not monotonic.

Does connecting a model to the internet remove hallucinations?

Partially. Real-time web access helps with current events and recent facts, but it doesn't solve the core problem: the model still invents details, misinterprets search results, or hallucinates about content it didn't actually retrieve. Internet access is a tool that reduces one class of hallucinations, not a cure.

How can I quickly check whether an answer is hallucinated?

Check sources: Click URLs or search for cited papers. If they don't exist, the answer is hallucinated. Verify facts: Check dates, names, and numbers against trusted sources. Compare multiple models: Ask the same question to different models. Sharp disagreement suggests at least one is hallucinating. Apply domain expertise: If you know the field, read critically for subtle implausibility—hallucinations often have tell-tale signs to expert readers.

Should I stop using AI if it hallucinates?

No. Use AI strategically with verification. For brainstorming, drafting, and exploratory work, hallucinations are a minor annoyance. For fact-critical work (research, compliance, medical decisions, financial advice), use AI as a starting point, then verify everything with trusted sources or expert review.

What's the difference between a hallucination and a genuine mistake?

A hallucination is confident and false. If a model says "I'm not sure, but it might be X," that's honest uncertainty, not hallucination. If it says "The capital of France is Berlin" with full confidence, that's a hallucination—the model stated something false as if it were fact. The hallmark is confident assertion of something untrue.

Sources

Wei, J., Wang, X., Schuurmans, D., et al. (2022). "Chain-of-Thought Prompting Elicits Reasoning in Large Language Models." ArXiv — the foundational paper demonstrating that step-by-step reasoning reduces hallucinations in math and logic tasks.
Maynez, J., Narayan, S., Hashimoto, B., & Hardt, D. (2021). "On Faithfulness and Factuality in Abstractive Summarization." ACL Proceedings — empirical study of hallucination rates and mechanisms in neural text generation.
Anthropic (2024). "Constitutional AI." https://www.anthropic.com/constitutional-ai — Anthropic's approach to reducing harmful outputs and hallucinations through principles-based training.