AI Content Detector
Check text for common patterns associated with AI-generated writing. Free, instant.
0 words
AI writing tools have gotten good enough that spotting their output by eye alone isn't always reliable — but certain statistical patterns still show up more often in AI-generated text than in typical human writing: unusually uniform sentence lengths, a cluster of stock transition phrases ('moreover', 'it's important to note', 'in conclusion'), and lower vocabulary variety across a passage.
This tool measures those three signals directly from the text you paste — sentence-length consistency, how many common AI-associated phrases appear, and how repetitive the vocabulary is — and combines them into a rough overall score. It does not connect to any AI provider's systems or detect a 'fingerprint' left by a specific model; it's pattern-matching against writing style, nothing more.
Because of that, treat the result as a talking point, not a verdict. Many people naturally write with consistent sentence structure and a predictable vocabulary, especially in formal or technical writing, which can trigger a higher score despite being entirely human-written. Likewise, heavily edited AI text can score low. This is genuinely useful for a quick gut-check, not for accusations.
Step-by-step instructions
- 1
Paste the text you want to check
At least 40 words and 3 sentences are needed for a meaningful reading.
- 2
Review the overall signal
See whether the text leans toward typical human-writing patterns or common AI-writing patterns.
- 3
Check the breakdown
Look at sentence uniformity, AI-phrase count, and vocabulary variety individually — a high overall score driven by just one factor is worth treating with extra skepticism.
Common use cases
A quick gut-check on a draft
Get a rough read on whether a piece of writing leans toward common AI patterns before deciding whether to dig deeper.
Reviewing your own AI-assisted writing
Check how strongly a passage you wrote with AI help still reads as AI-typical, before editing it further.
Spotting common AI phrasing habits
Learn to recognize the stock transition phrases and structural patterns that show up disproportionately in AI-generated text.
Tips for best results
- Never treat a single score as proof — use it as a starting point for a closer read, not a final answer.
- Short passages (under ~40 words) don't give the statistics enough to work with, which is why the tool asks for more text first.
- Formal, technical, or academic human writing can score higher than casual writing simply because it naturally uses more consistent sentence structure and domain vocabulary.
- If you're evaluating someone else's work for something high-stakes (academic integrity, hiring, publishing), combine this with other evidence and human judgment rather than relying on any single detector.
Frequently asked questions
How accurate is this tool?
It measures real statistical patterns correlated with AI-generated text, but it's a heuristic, not a certified detector — no tool, including commercial ones, can verify authorship with certainty. False positives and false negatives are both common with any AI detector.
Can this tell me which AI model wrote something?
No — it only measures general writing-style statistics, not model-specific signatures. It can't identify ChatGPT vs. another AI, or distinguish AI text from unusually uniform human writing.
Why did my own human-written text score high?
Consistent sentence length and a focused vocabulary aren't unique to AI — technical writers, non-native speakers writing carefully, and formal documents often show similar patterns.
Is my text uploaded anywhere?
No — all analysis happens locally in your browser using JavaScript. Nothing you paste is sent to a server.
Should I use this to accuse someone of using AI?
No. This tool is meant as a rough, informal indicator. It should never be the sole basis for an academic, legal, or employment decision — combine it with other evidence and direct judgment.