SignalsSome formulaic patterns
Sentence-length variation32% variation
Human writing mixes short sentences with long ones. Machine drafts tend toward a uniform middle length. Higher variation reads more naturally.
Sentence lengths: 20, 8, 12, 14
Average sentence length13.5 words
Long average sentences make prose heavy regardless of who wrote it. Around 15-20 words reads comfortably; past 25 the reader starts working.
Overused vocabulary12 instances (222.2 per 1000 words)
A small set of words appears far more often in generated text than in human writing. Their density is the single most visible tell.
delve · leveraging · robust · seamless · crucial · myriad · facilitate · landscape · underscores · embark · holistic · comprehensive
Hedge phrases2 found
Phrases like "it is important to note that" add words without adding meaning. Cutting them shortens the text and sharpens it.
it is important to note · In today's fast-paced world
Em dash density0 dashes (0.0 per 1000 words)
Models punctuate with em dashes several times more often than people typing on a keyboard, where the character is awkward to reach.
Connective openers3 of 4 sentences
Sentences that begin Moreover, Furthermore, Additionally are connective throat-clearing. A few are fine; a pattern of them is a tell.
Moreover, leveraging robust frameworks is crucial for success. · Furthermore, this seamless approach will facilitate numerous outcomes … · Additionally, organisations should embark on a holistic transformation…
This is not an authorship verdict. No detector can reliably establish who or what wrote a text - OpenAI withdrew its own for low accuracy, and false positives fall hardest on people writing in a second language. These are measurements of the writing, not claims about the writer. Please do not use them to accuse anyone.
Low variation and heavy hedging are worth fixing whoever wrote the draft.Rewrite it with the humanizer
Category:ai
Mode:Offline
Version:1.0.0
Access:Free
人工智慧探測器可以告訴你什麼,不能告訴你什麼
該工具測量使散文被閱讀為機器生成的寫作模式,並將每篇散文及其背後的證據報告。它不會輸出模型編寫文字的百分比機會,因為該數字無法誠實生成,而且生成它的危害是真實的。
這就是人工智慧探測器的問題。它們是統計分類器,從表面特徵猜測作者身份,而且它們不夠可靠,以至於 OpenAI 撤回了自己的分類器,因為準確性較低。它們的誤報最容易落在用第二語言寫作的人身上,他們的散文明顯更規則,因此對於分類器來說看起來更像機器。看起來自信的「87% 人工智慧生成」被用來指責學生或拒絕自由工作者,而其背後沒有任何東西可以承載這種分量。任何向您顯示如此數字的工具都會邀請您對拋硬幣做出相應的決定。
可以誠實測量的是文字本身。句子長度的變化,通常稱為突發,是最強的單一訊號:人類寫作將三個單字的句子與三十單字的句子混合在一起,而機器草稿則穩定在統一的中間長度。詞彙密度計算文本達到模型過度使用的一小組單字的頻率。對沖密度計算諸如"值得注意的是"之類的短語,這些短語增加了長度而沒有意義。 Em 破折號率計算字元模型產生的頻率比打字的人高出數倍。連接開機者計算以「此外」或「此外」開頭的句子。
其中每一個都是關於寫作的事實,而不是關於作者的主張,並且每個都報告有產生它的匹配項,因此您可以檢查測量結果而不是信任它。這使得結果對於它實際上有益的東西很有用:編輯。低變化和沈重的對沖會使散文變得更糟,無論是誰寫的,兩者都是可以修復的。
用它來診斷您自己的草稿,查看文件的哪些部分讀作公式化,並了解模式是什麼。不要使用它,也不要讓任何人使用它來決定一個人是否說出了自己工作的真相。一切都在您的瀏覽器中運行;沒有任何內容被上傳、記錄或儲存。
如何檢查人工智慧模式的寫作
1
貼上文字
任何散文 - 草稿、文章、提交內容。它會保留在您的瀏覽器中並且永遠不會上傳。
2
讀取每個訊號
六種測量值,每種測量值都有其值、簡單的英語解釋以及產生它的匹配項。
3
檢查證據
每個訊號都列出了它實際匹配的內容。如果計數看起來錯誤,證據會告訴您原因。
4
修復值得修復的東西
低變化和沈重的對沖會讓寫它的人寫得更糟。將文字發送給人性化的人,或手工編輯。
它衡量什麼
句子長度的變化(突發性)
報告為變異係數,因此它不依賴絕對長度。集合中最強的單一訊號。
過度使用的詞彙密度
計算大約 40 個單字模型持續達到,每 1000 個單字標準化,因此短文本和長文本進行比較。
對沖和開盤點很重要
增加長度但沒有意義的短語,以及以連接清喉嚨開頭的句子,兩者都列出了匹配項。
每個數字的證據
每個訊號都顯示實際匹配或原始句子長度,因此可以檢查而不是相信測量結果。
Frequently Asked Questions
不,任何其他探測器也不能,無論其輸出看起來多麼自信。這衡量的是寫作模式,而不是作者身份。將一個給你一定百分比的工具視為猜測,因為它確實如此。
Related Tools
ATH
ai
AI Text Humanizer
Rewrite the habits that make prose read as machine-written - inflated vocabulary, hedge phrases, em dashes, and connective filler.
CWR
ai
Claude Watermark Remover & Detector
Detect and remove invisible Unicode characters from AI text - zero-width joiners, variation selectors, bidi controls, and tag characters.
RNG
generators
Random Number Generator
Generate random numbers in any range, with unique-only and decimal options.
WC
text
Word & Letter Counter
Count words, letters, sentences, paragraphs, and estimate reading time
Comments
0 comments
No comments yet. Be the first to share your thoughts!
ai writing checkerai content detectorai text detectorcheck if text is aiburstiness checkerai writing patternsperplexity checkerwriting style analyser