又是一年畢業(yè)季,在多個社交平臺上,關于論文AI率“亮紅燈”的帖子引發(fā)熱議。

有網(wǎng)友反映,畢業(yè)論文中原創(chuàng)內(nèi)容經(jīng)一些系統(tǒng)檢測后,常被指出AI生成內(nèi)容比例過高。

不少大學生分享了降低AI率的經(jīng)歷:有人反復修改結(jié)果導致語句生硬;有人被迫刪減精彩段落;更有甚者調(diào)侃需故意“寫笨”以規(guī)避誤判。
A new academic dilemma has emerged as Chinese university students report unexpected "high AI-generated content" alerts when using popular plagiarism detection systems to check their graduation theses that they wrote without AI assistance. Many have resorted to awkward phrasing modifications, deletion of quality paragraphs or even intentional "dumbing down" of language to bypass AI detection algorithms.
據(jù)大河報消息,將朱自清的名篇《荷塘月色》與劉慈欣《流浪地球》的片段上傳至某常用論文檢測系統(tǒng)后,結(jié)果顯示,這兩篇經(jīng)典作品AI生成內(nèi)容總體疑似度竟分別達到了62.88%和52.88%。
In a startling experiment reported by media, excerpts from Zhu Ziqing's literary classic Moonlight Over the Lotus Pond and Liu Cixin's science fiction The Wandering Earth were submitted to a commonly used thesis detection system. The results showed 62.88 percent and 52.88 percentAI-generated content suspicion rates, respectively, for these two authors' original works.

面對這樣的檢測數(shù)據(jù),不少網(wǎng)友擔心自己的論文也會被誤判AI率過高,從而影響畢業(yè)。

多所高校檢測畢業(yè)論文AI率
有大學生花錢“求通過”
據(jù)了解,多所高校曾發(fā)文對畢業(yè)論文AI生成內(nèi)容進行規(guī)范。
面對AI率檢測的壓力,有大學生開始向聲稱可以提供“降AI率”服務的店家尋求幫助。

這些服務的常用手法包括替換同義詞、調(diào)整句子順序,或加入不必要的修飾和口語化表達。部分商家甚至會先用自動化工具初步修改,再少量人工調(diào)整。
但這種“修改”往往使論文語言生硬,學術(shù)性降低,對降低AI率效果有限,有時反而弄巧成拙。更嚴重的是,學生“花錢包過”,還可能面臨詐騙和隱私泄露風險。
Meanwhile, a shadow market has emerged offering "AI-reduction" services ranging from mechanical synonym swaps to artificial colloquialization. However, these quick fixes often degrade academic rigor while exposing users to fraud and data breaches.

專家提醒:AI論文反識別或存在誤差
“當前AI率檢測是AI技術(shù)融入學術(shù)領域的初期表現(xiàn),其標準與技術(shù)均在發(fā)展和完善過程中?!敝心洗髮W講師關耳(化名)分析道。她認為,各高校的探索實踐也反映出,在學術(shù)層面,一個普遍適用的指導性標準尚未完全形成。
“除了標準不一樣讓人頭疼,AI檢測技術(shù)本身到底準不準,也是個大問題。畢竟,這直接關系到學生辛辛苦苦寫的原創(chuàng)內(nèi)容,會不會被機器錯怪成AI寫的?!?關耳覺得“目前的AI檢測技術(shù)用起來還有些問題”,給出的結(jié)果最好只當個參考,不能完全以此來判斷一篇論文的好壞。
"Current detection tools represent early-stage attempts to integrate AI governance into academia, not definitive quality assessments," cautioned Central South University lecturer Guan Er (pseudonym). She noted overreliance on flawed detection metrics risks distorting academic values. Unlike clear-cut plagiarism cases, most institutions currently treat high AI scores as teaching opportunities rather than misconduct, reflecting awareness of the technology's limitations.



據(jù)央視新聞此前報道,目前國內(nèi)有多個團隊正在從不同方向發(fā)力,進行AI論文反識別的檢測研究。一些學生和專家提醒,AI工具是根據(jù)它捕獲的海量信息進行的智能分析,但這些信息的真實性和準確性可能存在缺陷,從而得出錯誤的結(jié)論。
關耳強調(diào),一篇論文怎么樣,關鍵還是看學生有沒有自己的思考和原創(chuàng)的東西。
As multiple research teams develop next-generation detection systems, educators emphasize cultivating authentic critical thinking. "The essence of scholarship remains human insight," Guan said.
據(jù)她觀察,目前多數(shù)高校對于AI生成內(nèi)容比例超標的情況,處理方式與對待查重率超標這類明確的學術(shù)不端問題尚有不同,更多的是引導修改,而非直接等同于《學位法》中嚴厲處罰的代寫、剽竊行為?!斑@種審慎的處理,也說明學術(shù)界已認識到AI檢測技術(shù)的工作原理存在一定局限性”。
堵不如疏,專家看來,強化師生的AI素養(yǎng)非常重要。近兩年,眾多高校都把人工智能作為通識必修課,不斷推動師生AI素養(yǎng)的提升。
來源:中國青年報
熱門跟貼