Master the techniques for identifying AI-generated content across writing, images, and multimedia. These prompts guide you through linguistic analysis, fact-checking methodologies, and advanced detection patterns used by professionals and educators.
You are an AI detection expert. Analyze the following text and identify patterns that suggest AI generation. Look for: repetitive phrasing, unnatural transitions, generic examples, mathematical precision, lack of personal voice, and systematic structure. Text: [INSERT TEXT] Provide a confidence score (0-100%) and explain your reasoning.
Compare these two writing samples side by side. Identify differences in vocabulary diversity, sentence structure complexity, emotional authenticity, and idiomatic language use. Determine which is more likely AI-generated and why. Sample 1: [TEXT] Sample 2: [TEXT]
Extract the following from this text: (1) Unique personal experiences mentioned, (2) Contradictions or inconsistencies, (3) Contextual knowledge that seems out of place, (4) Stylistic patterns. Use these to assess likelihood of AI generation. Text: [INSERT TEXT]
Analyze writing for signs of temperature/creativity settings. Look for: high creativity (unexpected metaphors, unusual word choices), medium (balanced style), low creativity (formulaic, repetitive). What does this suggest about the source? Text: [INSERT TEXT]
Identify whether this text shows signs of prompt engineering or fine-tuning. Look for: sudden tone shifts, template-like structures, repeated section headers, artificial transitions. Rate the likelihood: High/Medium/Low. Text: [INSERT TEXT]
Create a detection checklist for evaluating [SPECIFIC CONTENT TYPE]. Include linguistic markers, structural patterns, contextual anomalies, and stylistic red flags specific to this format.
Perform a readability analysis on this text. Calculate Flesch Reading Ease, identify clause complexity, assess vocabulary level. Compare to human writing benchmarks. Does the pattern suggest AI authorship? Text: [INSERT TEXT]
Evaluate this text for passive voice usage, adverb density, and connector word frequency. AI writing often has predictable patterns in these metrics. Analyze and rate likelihood of AI generation. Text: [INSERT TEXT]
Identify all instances of hedging language (perhaps, might, may, arguably). What's the density? AI models tend toward specific hedging patterns. Does this text match those patterns? Text: [INSERT TEXT]
Extract sentences with perfect grammar but awkward phrasing. These often indicate AI generation attempting to follow rules too rigidly. List examples and explain why they feel unnatural. Text: [INSERT TEXT]
Analyze the coherence chains in this text. Do pronouns clearly reference antecedents? Are topic shifts explicit or jarring? AI detection often relies on these linguistic patterns. Text: [INSERT TEXT]
Perform a sentiment analysis on each paragraph. AI writing often shows unnatural sentiment consistency. Map sentiment trajectory and identify anomalies. Text: [INSERT TEXT]
Fact-check specific claims in this text: [LIST 5 CLAIMS]. For each: (1) Search for evidence, (2) Check reliability of sources, (3) Identify if the claim could be AI hallucination, (4) Provide verification status.
Examine citations and references in this text. Are they: (1) Real and correctly attributed, (2) Partially real but misquoted, (3) Fabricated (hallucinated). List findings. Text: [INSERT TEXT]
Analyze this text for temporal inconsistencies, outdated information, or anachronistic references. Would a human expert in this field make these errors? What does this suggest? Text: [INSERT TEXT]
Verify specific statistics, percentages, and data points mentioned in this text. Are they accurate? Are sources cited? Do numbers match official records? Text: [INSERT TEXT]
Check for self-contradictions within this text. Does the author support position X in paragraph 1 but contradict it in paragraph 5? Humans usually avoid this; AI sometimes exhibits inconsistency. Text: [INSERT TEXT]
Investigate any expert or authority quotes in this text. Are they: (1) Accurately quoted, (2) Real people/publications, (3) Contextually appropriate. Document your findings. Text: [INSERT TEXT]
Analyze the use of examples in this text. Are they: (1) Concrete and specific, (2) Generic/illustrative, (3) Diverse or repetitive. AI often uses templated examples. Assessment? Text: [INSERT TEXT]
Evaluate metaphor and analogy use. Are they: (1) Original and creative, (2) Common/clichéd, (3) Mixed in appropriateness. What does this suggest about authorship? Text: [INSERT TEXT]
Analyze humor usage, if present. Is it: (1) Contextual and natural, (2) Forced or out of place, (3) Reliant on obvious joke structures. Assessment? Text: [INSERT TEXT]
Examine the opening and closing of this text. Do they: (1) Hook the reader naturally, (2) Feel formulaic (we'll explore, in conclusion), (3) Show personality. What does this indicate? Text: [INSERT TEXT]
Rate the vocabulary range. Does the author use: (1) Varied vocabulary with appropriate complexity, (2) Limited repetitive vocabulary, (3) Overly complex but generic words. What does this suggest? Text: [INSERT TEXT]
Analyze narrative voice consistency. Does the author maintain the same perspective, tone, and personality throughout? Shifts might indicate: (1) Multiple authors, (2) AI generation with temperature changes, (3) Human inconsistency. Assessment? Text: [INSERT TEXT]
Perform a token prediction analysis. In human writing, which words would a native speaker find predictable vs. surprising? If the text matches predictable token patterns consistently, it might be AI. Text: [INSERT TEXT]
Analyze the text's semantic coherence using word embeddings mentally. Do concepts logically progress or feel randomly associated? AI sometimes shows subtle incoherence. Text: [INSERT TEXT]
Evaluate complexity curves: Do ideas build progressively or restart at each paragraph? AI often resets complexity. Map this pattern. Text: [INSERT TEXT]
Assess persona consistency. If the author claims expertise/experience, does language reflect that? Mismatches often indicate: (1) AI claiming false expertise, (2) Content scraping, (3) Ghostwriting. Text: [INSERT TEXT]
Analyze information density per sentence. Is it: (1) Varied (human), (2) Consistently high/low (AI default), (3) Strategic. Pattern suggests? Text: [INSERT TEXT]
Perform background knowledge inference. What would a human with the claimed expertise know that this text ignores? Gaps can indicate: (1) Insufficient human knowledge, (2) AI training data limitations, (3) Intentional simplification. Text: [INSERT TEXT]
Need custom AI detection prompts for your specific use case?
Get Custom Prompts