顯示具有 文圖學 標籤的文章。 顯示所有文章
顯示具有 文圖學 標籤的文章。 顯示所有文章

2026/08/05

汇流之外 Beyond Crosscurrents




 

走进任何一个关于东西文化、丝绸之路、美术工艺的文物展览,必然遇见"交流""融合""文明互鉴"这类词语。文物本来就是混杂的,工艺本来就是流动的,艺术从来不在封闭的容器里生长——这些都是实存的,也都不奇特。那么,一个以"Crosscurrents"(汇流)为名的展览,还能告诉我们什么?

在新加坡亚洲文明博物馆(Asian Civilisations MuseumACM)的《汇流:卢浮宫的莫卧儿、萨法维与奥斯曼艺术珍品》(Crosscurrents: Masterpieces of Mughal, Safavid and Ottoman Art from the Musée du Louvre)特展,第一眼看到的是一只十六世纪奥斯曼帝国的玉杯。

通体沉稳的黑绿色,杯身以金丝嵌出繁复纹饰,工匠在玉面凿出浅槽,以两种镶嵌技法填入金丝,制造出层次感。花形开光中,框住每一粒红宝石。开光之间,蔓延着卷叶纹。展签记录它曾属于法国国王路易十四,收藏于凡尔赛宫,后来进入卢浮宫。

我的视线在展签最后一行停下来:红宝石经近年分析,认为来自缅甸。

缅甸?原来和东南亚有缘分啊。

历史上,亚美尼亚商人掌握着缅甸红宝石的主要贸易路线,将宝石通过孟加拉湾,进入莫卧儿帝国(印度),萨法维(伊朗)与奥斯曼(土耳其)。这个玉杯在展场开宗明义,呈现了汇流的主题。

细看杯身的花形开光与卷叶纹。这套图式语言贯穿大半个展览:抄本的装饰页、书页的题花、清真寺的门拱轮廓。

我想起去年底在上海浦东美术馆举办的同类展览《图案的奇迹》(The Wonder of Patterns: Masterpieces from India, Iran and the Ottoman World from the Musée du Louvre),以约300件卢浮宫藏品呈现伊斯兰艺术的美学成就。ACM的策展团队将自身的30件藏品加入展览,与卢浮宫的100件借展文物并置陈列,于是,故事的视角从西方转向了东南亚。

策展,就是一种立场和价值判断。

波斯对马来文学的影响(Persian Influence on Malay Letters)展区里,并排陈列了奥斯曼土耳其(1752年)与马来西亚丁加奴(18-19世纪)两部《古兰经》(Quran)。奥斯曼的版本装帧精整,是便于记诵的标准形制;丁加奴的版本在借鉴这套版式的同时,发展出一种本地独特的"花卉书法"floral calligraphy——阿拉伯文字母在抄写者的笔下,逐渐转化为植物的形态,字母与花草之间的边界被有意模糊。我们还看到19世纪的马来亚《圣纪颂》(Kitab Mawlid)也印证了这一点:书页上方的尖拱题花与卷叶藤蔓,与玉杯杯身的开光共享同一套图式语言——只是载体从金玉变成了纸页。

在文图学(Text and Image Studies)的视角下,这构成了一个具体的论题:卷叶纹与拱形开光,玉器和抄本,卷叶纹更进一步渗入文字本身,成为书法的一部分。奥斯曼的书法彰显文字的神圣权威;丁加奴的花卉书法,是文字在图像化过程中找到的另一种尊严——而这种秩序,在东南亚的语境里生长出来。

另一个展区里,一座来自苏门答腊西部的清真寺模型也吸引了我的注意。八角形开放式柱廊撑起宽大的多层坡屋顶,上层收为较小的亭台,顶端是一个洋葱形圆顶,收尖为塔刹。坡屋顶是东南亚本地的建筑传统;洋葱形圆顶则是后来引入的伊斯兰形式。两者并存于同一座建筑里,不是折衷,而是一个社群主动选择的视觉表达:我们在宽阔的伊斯兰文明之中,同时也在这片土地上。

汇流,当然发生了。但汇流之后,接收的一方做了什么,才是更值得细思的问题。丁加奴的抄写者把字母写成花草,苏门答腊的清真寺模型建造者在洋葱顶下保留了坡屋顶,这些都不是被动接收的痕迹,而是在地的主体性(local agency)。

现场展柜一框接一框、主题接主题。它能并置,能提示,但持续编辑与创造,最终的连贯性,必须由每一个观众去完成。

离开展厅前,我再一次走回那只玉杯前。黑绿色的玉面在聚光灯下透出沉稳的光泽,描金深浅有致那些宝石的颜色,随着我的目光偏移,或鲜红、或艳紫。500年,东南亚早就在这件文物里了。只是需要站在这里,仔细端详,反映自我,才能看见。

汇流,是过去发生的事。汇流之外,是今天的观者,带着自己的位置与记忆,重新读出文物里那些未被命名的在场。

 

202681日,新加坡《联合早报》上善若水专栏

 

Beyond Crosscurrents

I Lo-fen

Walk into almost any exhibition of cultural relics devoted to East–West encounters, the Silk Road, or arts and crafts, and one is bound to encounter terms such as “exchange,” “fusion,” and “mutual learning among civilisations.” Cultural objects have always been hybrid; craftsmanship has always travelled; art has never grown inside a sealed container. All of this is historically real, and none of it is especially surprising. What, then, can an exhibition entitled “Crosscurrents” still tell us?

At the Asian Civilisations Museum (ACM) in Singapore, the special exhibition Crosscurrents: Masterpieces of Mughal, Safavid and Ottoman Art from the Musée du Louvre opens with a sixteenth-century Ottoman jade cup.

The vessel is a composed, deep blackish green. Intricate patterns are inlaid across its surface in gold wire. The craftsman carved shallow grooves into the jade and filled them with gold using two different inlay techniques, creating a sense of depth and layering. Each ruby is framed within a floral cartouche, while scrolling foliage spreads between the cartouches. The label records that the cup once belonged to King Louis XIV of France, was kept at the Palace of Versailles, and later entered the Louvre.

My eyes paused at the final line of the label: recent analysis suggests that the rubies came from Myanmar.

Myanmar? So the cup had a connection with Southeast Asia after all.

Historically, Armenian merchants controlled the principal trade routes for Burmese rubies, transporting the gemstones across the Bay of Bengal into the Mughal Empire in India, Safavid Iran, and the Ottoman Empire in Türkiye. Placed at the beginning of the exhibition, this jade cup announces the theme of crosscurrents from the outset.

A closer look at the floral cartouches and scrolling foliage on the cup reveals a visual language that runs through much of the exhibition: the decorated opening pages of manuscripts, ornamental headings on book pages, and the outlines of mosque arches.

I was reminded of a related exhibition held late last year at the Museum of Art Pudong in Shanghai, The Wonder of Patterns: Masterpieces from India, Iran and the Ottoman World from the Musée du Louvre, which presented the aesthetic achievements of Islamic art through some 300 objects from the Louvre. The ACM curatorial team added thirty works from its own collection and displayed them alongside one hundred objects on loan from the Louvre. In doing so, the perspective of the story shifted from the West towards Southeast Asia.

Curating is itself a position and a value judgement.

In the section “Persian Influence on Malay Letters,” two Qurans are displayed side by side: one from Ottoman Türkiye, dated 1752, and the other from Terengganu, Malaysia, dating from the eighteenth to nineteenth century. The Ottoman volume is precisely bound and follows a standard format designed to facilitate memorisation. While adopting this layout, the Terengganu version developed a distinctive local form of “floral calligraphy.” In the hands of the scribe, Arabic letters gradually transform into vegetal forms, deliberately blurring the boundary between script and flowers.

A nineteenth-century Malay Kitab Mawlid further demonstrates this development. The pointed-arch heading and scrolling vines at the top of the page share the same visual language as the cartouches on the jade cup, except that the medium has shifted from gold and jade to paper.

From the perspective of Text and Image Studies, this presents a specific line of inquiry: scrolling foliage and arched cartouches move between jade vessels and manuscripts, while vegetal ornament penetrates even further into the written word itself, becoming part of calligraphy. Ottoman calligraphy proclaims the sacred authority of the text. Terengganu floral calligraphy gives writing another form of dignity through its transformation into image—an order that took shape within the context of Southeast Asia.

In another section, a model of a mosque from western Sumatra also caught my attention. An octagonal open colonnade supports a broad, multi-tiered pitched roof. The upper level narrows into a smaller pavilion, topped by an onion-shaped dome that tapers into a finial. The pitched roof belongs to the local architectural tradition of Southeast Asia, while the onion dome is a later-introduced Islamic form. Their coexistence within a single structure is not a compromise, but a visual statement actively chosen by a community: we belong within the vast world of Islamic civilisation, and we also belong to this land.

Crosscurrents undoubtedly occurred. Yet what the receiving culture did after those currents arrived is the more important question. The scribes of Terengganu transformed letters into flowers and plants; the builders represented by the Sumatran mosque model preserved the pitched roof beneath the onion dome. These are not traces of passive reception, but expressions of local agency.

In the gallery, one display case follows another, one theme follows the next. The exhibition can juxtapose and suggest, but the continuity of ongoing adaptation and creation must ultimately be completed by each individual visitor.

Before leaving the gallery, I returned once more to the jade cup. Under the spotlight, the blackish-green jade emitted a restrained lustre, while the varying depths of the gold inlay created a subtle rhythm. As my gaze shifted, the gemstones appeared at times bright red, at times vivid purple. For five hundred years, Southeast Asia had already been present within this object. It simply required someone to stand before it, look carefully, reflect upon one’s own position, and finally see it.

Crosscurrents belong to what happened in the past. Beyond crosscurrents stands the viewer of today, carrying her own position and memories, rereading within these objects the presences that have never been named.

1 August 2026, “Shang Shan Ruo Shui” column, Lianhe Zaobao, Singapore.

 

2026/07/04

终于可以靠近壁画了 Finally, We Can Get Close to the Murals


 第一次真正靠近中国壁画,是在山西晋中的晋之源壁画艺术博物馆。

靠近,并不是因为我第一次看见壁画。事实上,多年来,我去过敦煌、新疆克孜尔、山西大同,也在不同地方看过不少寺观壁画。真正的问题是:过去虽然看见了,却未必真的看清

洞窟太暗,距离太远,时间太短,许多壁画又高悬于穹顶或梁壁之上。人站在现场,往往只能仰头匆匆一瞥。知道那里有飞天,有菩萨,有山水,有人物,却难以真正进入那些线条、色彩与细节之中。

晋之源壁画艺术博物馆展示的,并不是原作壁画,而是经过高清扫描后的数字输出图像。它以中国壁画史为脉络,将不同朝代、不同地区的经典壁画重新组织成一个可观看、可阅读、可比较的视觉空间。

有趣的是,正因为它不是原作,我反而第一次真正看见了许多过去没看见的东西。

我开始看见人物衣纹的转折,看见矿物颜料留下的层次,看见墙面裂痕之间仍未消失的笔触,看见飞天的眼神,看见永乐宫神仙队伍里不同人物的表情差异。我甚至第一次意识到,中国壁画原来有如此强烈的运动感

过去在寺观或洞窟里,人们往往被一种朝圣式的观看方式所限制。人必须顺着空间移动,顺着光线移动,也顺着时间移动。壁画属于宗教空间的一部分,而不是单独存在的艺术对象。

但进入博物馆之后,壁画的观看逻辑改变了。它们从宗教现场进入展览空间,从建筑的一部分变成艺术史的一部分。

这让我不断思考一个问题:我们今天看到的,究竟还是壁画吗?或者说,它已经是一种经过技术输出转译后的新文本

过去我谈的文图学,讨论的是各种形式的文本生产与诠释,而到了AIGC时代,我们开始面对另一种新的情况:任何图像,都可能经过扫描、计算、数据化与再生成。

于是,图像不再只是图像,它同时也是数据。

晋之源壁画艺术博物馆给我的震撼,恰恰正在这里。

我发现,数字技术并不只是复制壁画,而是在重新组织人与壁画之间的关系。过去,壁画的重点在保存;今天,壁画开始进入传播。过去,壁画强调原址;今天,壁画开始脱离原址,进入新的媒介环境。过去,观看是一种空间经验;今天,观看同时也是一种数据经验。

这些,其实都与AIGC时代的人类视觉经验非常接近。

现在许多人第一次看见艺术作品,并不是在美术馆,而是在手机屏幕上。甚至有些年轻人,是先看见AI生成的敦煌风”“宋画风,才开始对传统艺术产生兴趣。

有人因此担忧:数字化会不会让人越来越不在乎原作?但我在晋之源壁画艺术博物馆里,反而产生相反的感受。

正因为数字扫描让我看清了细节,我才更想去看原作。原作与数字图像之间,并不是取代关系。数字图像提供的是可阅读性,原作提供的则是存在感。高清扫描能够让人看见裂痕,却无法完全传达墙壁的厚度;能够让人看见线条,却无法替代人在空间中的身体感;能够让人看清颜色,却无法复制时间真正沉积于墙面上的气息。

真正的壁画,仍然带着一种时间重量。但数字技术却让更多人终于有机会靠近它。长期以来,中国壁画始终处于一种看得见却不容易靠近的状态。与卷轴绘画相比,壁画太巨大,也太遥远。它们存在于山中、洞窟中、寺庙中,甚至存在于普通人难以抵达的地方。某种意义上,中国壁画一直是一种被遮蔽的视觉传统

然而今天,高清扫描、数字输出、人工智能图像处理技术,却开始改变这种状况。它们让壁画重新进入大众视野,也重新进入当代人的视觉经验。晋之源壁画艺术博物馆,正好提供了一个非常具体的现场。

在那里,我忽然意识到:我们终于可以靠近壁画了。

但更重要的是——当我们终于能够靠近壁画时,我们也正在重新学习,如何观看。

 

202674日,新加坡《联合早报》“上善若水”专栏

 

Finally, We Can Get Close to the Murals

I Lo-fen

The first time I truly came “close” to Chinese murals was at the Jinzhiyuan Mural Art Museum in Jinzhong, Shanxi.

By “close,” I do not mean that it was the first time I had ever seen murals. In fact, over the years, I have been to Dunhuang, Kizil in Xinjiang, and Datong in Shanxi, and I have also seen many temple and monastery murals in different places. The real issue is this: although I had seen them before, I had not necessarily seen them clearly.

The caves were too dark, the distance too great, and the time too short. Many murals were also high above, on domes or beams and walls. Standing on site, one often could only look up and cast a hurried glance. One knew that there were flying apsaras, bodhisattvas, landscapes, and figures there, but it was difficult to truly enter into those lines, colours, and details.

What the Jinzhiyuan Mural Art Museum displays are not original murals, but digitally printed images produced after high-definition scanning. Taking “the history of Chinese murals” as its thread, it reorganizes classic murals from different dynasties and different regions into a visual space that can be viewed, read, and compared.

Interestingly, precisely because they are not the originals, I for the first time truly saw many things that I had not seen before.

I began to see the turns of the figures’ garment folds, the layers left by mineral pigments, the brushstrokes that had not yet disappeared among the cracks on the wall surface, the eyes of the flying apsaras, and the differences in expression among the various figures in the procession of immortals at Yongle Palace. I even realized for the first time that Chinese murals possess such a powerful sense of movement.

In the past, in temples, monasteries, or caves, people were often restricted by a kind of “pilgrimage-style” mode of viewing. One had to move according to the space, according to the light, and also according to time. Murals belonged to a religious space; they were not independent art objects.

After entering the museum, however, the logic of viewing the murals changed. They moved from religious sites into exhibition spaces, and from parts of architecture into parts of art history.

This made me keep thinking about one question: what we see today—are they still “murals”? Or have they already become a kind of “new text” translated through technological output?

In the past, the Text and Image Studies I discussed concerned the production and interpretation of texts in various forms. But in the age of AIGC, we have begun to face another new situation: any image may undergo scanning, computation, datafication, and regeneration.

Thus, an image is no longer merely an image. It is also data.

The shock that the Jinzhiyuan Mural Art Museum gave me lies precisely here.

I discovered that digital technology is not merely “copying” murals, but reorganizing the relationship between people and murals. In the past, the emphasis of murals lay in “preservation”; today, murals have begun to enter “dissemination.” In the past, murals emphasized the original site; today, they have begun to detach from the original site and enter new media environments. In the past, viewing was a spatial experience; today, viewing is also a data experience.

All of these are in fact very close to human visual experience in the age of AIGC.

Today, many people first encounter works of art not in art museums, but on mobile-phone screens. Some young people even first see AI-generated “Dunhuang style” or “Song painting style” images before they begin to take an interest in traditional art.

Some people therefore worry: will digitization make people care less and less about the originals? But in the Jinzhiyuan Mural Art Museum, I had the opposite feeling.

Precisely because digital scanning allowed me to see the details clearly, I wanted even more to see the originals. The relationship between originals and digital images is not one of replacement. What digital images provide is “readability”; what originals provide is “presence.” High-definition scanning can allow people to see cracks, but it cannot fully convey the thickness of a wall; it can allow people to see lines, but it cannot replace the bodily sensation of being in a space; it can allow people to see colours clearly, but it cannot reproduce the atmosphere of time truly accumulated on the wall surface.

True murals still carry a certain weight of time. Yet digital technology has allowed more people finally to have the chance to “come close” to them. For a long time, Chinese murals have remained in a state of being “visible yet not easy to approach.” Compared with scroll paintings, murals are too vast and too distant. They exist in mountains, caves, temples, and even in places difficult for ordinary people to reach. In a certain sense, Chinese murals have always been a “concealed visual tradition.”

Today, however, high-definition scanning, digital output, and artificial intelligence image-processing technologies have begun to change this situation. They allow murals to re-enter public view and to re-enter the visual experience of contemporary people. The Jinzhiyuan Mural Art Museum provides a very concrete site for this.

There, I suddenly realized: we can finally get close to murals.

More importantly—when we can finally get close to murals, we are also relearning how to look.

4 July 2026, Singapore, Lianhe Zaobao, “Shangshan Ruoshui” column

 

2026/06/19

苏东坡《浣溪沙·端午》千年前的端午,原来如此优雅 Su Shi's Dragon Boat Festival: Memory Across a Thousand Years

 






《浣溪沙·端午》


宋·苏轼


轻汗微微透碧纨,

明朝端午浴芳兰。

流香涨腻满晴川。


彩线轻缠红玉臂,

小符斜挂绿云鬟。

佳人相见一千年。


白话译文


薄薄的丝衣上微微透出香汗,

明天就是端午节了,人们将用兰草浸泡的香汤沐浴。

芬芳的香气弥漫四野,仿佛充满了整个晴朗的河川。


女子的手臂上轻轻缠着五彩丝线,

发髻旁斜挂着驱邪祈福的符袋。

愿这样美好的佳节与美丽的人儿,千年之后依然能够相见。


English Translation

Huan Xi Sha: The Dragon Boat Festival


A light sheen of fragrance moistens her silk attire;

Tomorrow, at the Dragon Boat Festival,

People bathe in orchid-scented water.


Sweet aromas drift and spread,

Filling the bright riverbanks with lingering perfume.


Colored threads gently circle her jade-like arms;

A talisman hangs beside her cloud-like hair.

May such beauty and this festive joy

Meet again a thousand years from now.

2026/05/11

你真的会用 AI 做学术研究吗?Do You Really Know How to Use AI for Academic Research?


 你现在怎样使用 AI?


是还在观望?

是已经开始使用,却不知道边界在哪里?

是担心 AI 影响原创性、学术伦理与研究判断?

还是已经想把 AI 纳入自己的研究方法?


我设计了一个 AIGC 学术人格测试,帮助你看见自己在 AIGC 时代的研究状态。


10 题测出你属于哪一种类型:

观望者、适用者、焦虑者,还是方法建构者。


AI 工具会更新,方法带你走对方向。


如果你发现自己正需要 AI 使用原则、学术伦理边界、披露模板与研究方法,欢迎预购我的新书:


《AIGC 时代的人文学术研究方法》

Humanities Research Methods in the Age of AIGC

作者:衣若芬 I Lo-fen


本书不是一本 AI 工具清单,而是帮助你建立 AIGC 时代人文学术研究方法的书。


第一阶段预购同步进行中。

完成付款,才视为预购成功。


测试链接:https://tally.so/r/GxY8eO


预购链接1:https://forms.gle/doqk2E2gKyEEaeZb6

预购链接 2:https://p.baominggongju.com/share?eid=69f993cdc1e7273224f92119



How are you using AI in your academic work?


Are you still observing from a distance?

Have you started using AI but remain unsure about its boundaries?

Are you worried about originality, academic integrity, and research judgment?

Or are you ready to integrate AI into your own research methodology?


I designed the AIGC Academic Persona Test to help you understand your research mindset in the age of AIGC.


In 10 questions, discover which type you are:

the Observer, the Practitioner, the Anxious User, or the Method Builder.


AI tools evolve. Method leads the way.


If you find yourself needing clearer principles, ethical boundaries, disclosure templates, and research methods for using AI in academic work, you are welcome to pre-order my new book:


Humanities Research Methods in the Age of AIGC

《AIGC 时代的人文学术研究方法》

Author: I Lo-fen 衣若芬


This is not a book of AI tool lists.

It is a book about building humanities research methods in the age of AIGC.


The first-stage pre-order is now open.

A pre-order is confirmed only after payment is completed.


Test link: https://tally.so/r/GxY8eO


Pre-order link 1: https://forms.gle/doqk2E2gKyEEaeZb6

Pre-order link 2:  https://p.baominggongju.com/share?eid=69f993cdc1e7273224f92119


2026/05/09

AI 工具会更新,方法带你走对方向

 


感谢踊跃预购!

《AIGC 时代的人文学术研究方法》第一阶段预购即将截止!

预购单

AI 工具越来越多,可是人文学术研究真正需要的,不是追着每一个工具跑。

工具会过时,方法不会。

在 AIGC 时代,真正重要的不是会用多少 AI,而是知道怎样用、何时用、用到哪里为止。

这本书不是一本 AI 工具清单,而是从人文学术研究的实际现场出发,讨论如何在论文写作、文献整理、资料分析、课堂教学、学术发表与研究伦理中,建立一套可以判断、可以说明、可以负责的 AI 使用原则与研究方法。

书中提供多种可以立即使用的模板、清单与表格,包括:

AIGC 使用披露声明模板、学术诚信自查清单、三大国际相关学术伦理规范体系核心要点对照表、人机协作写作工作流、推荐提示词示例、AI 使用日志模板、常用数据库与检索路径速查表、注脚格式示例、伦理申请基本清单,以及投稿、同行评审与预发表流程。

你不需要记住所有 AI 工具。

你真正需要的是:

知道什么时候可以用 AI,什么时候不该用;

知道 AI 可以帮你做到哪一步,哪一步必须由自己判断;

知道怎样说明自己的 AI 协作过程;

知道如何在提高效率的同时,守住学术伦理与研究主体性。

完成付款,才视为预购成功。

只登记但未付款者,不列入第一阶段预购名单。

已经登记但尚未付款的读者,请尽快完成付款并电邮付款资料。预购与付款方式请见表单说明。

2026/04/11

【文图学】【AIGC文图学】的定义 Definition of Text and Image Studies.Text and Image Studies on AIGC

 


【文图学】【AIGC文图学】的定义

引用来源:衣若芬《AIGC时代的人文学术研究方法》(2026)

I Lo-fen,Humanities Research Methods in the Age of AIGC,2026


文图学是由衣若芬提出的跨学科研究领域与方法,以广义文本观为基础,研究文字、图像及其他文本形态的关系、互动、张力与生成,探讨文本及其意义在媒介机制、社会网络、文化背景与历史语境中的生产、传递、转化与理解。

Text and Image Studies: an interdisciplinary field and method proposed by I Lo-fen, grounded in a broad concept of text. It examines the relationships, interactions, tensions, and generative processes among words, images, and other textual forms, and explores the production, transmission, transformation, and understanding of texts and their meanings within media mechanisms, social networks, cultural contexts, and historical circumstances.

AIGC 文图学:由衣若芬提出并发展的跨学科研究领域与方法,以广义文本观为基础,面对人工智能能够生成文字、图像、声音、影像等多种文本的新条件,研究文本如何形成、如何被理解与判断,并探讨不同文本形态的关系、机制、意义及其媒介条件、社会网络、文化背景与历史语境.

Text and Image Studies on AIGC: an interdisciplinary field and method proposed and developed by I Lo-fen. Grounded in a broad concept of text, it responds to the new condition in which artificial intelligence can generate multiple forms of text, including words, images, sound, and video. It studies how texts are formed, understood, and judged, and explores the relations, mechanisms, and meanings of different textual forms, as well as their media conditions, social networks, cultural contexts, and historical circumstances.


【文图学】【AIGC文图学】的定义 Definition of Text and Image Studies.Text and Image Studies on AIGC


 

【文图学】【AIGC文图学】的定义

引用来源:衣若芬《AIGC时代的人文学术研究方法》(2026)

I Lo-fen,Humanities Research Methods in the Age of AIGC,2026

文图学是由衣若芬提出的跨学科研究领域与方法,以广义文本观为基础,研究文字、图像及其他文本形态的关系、互动、张力与生成,探讨文本及其意义在媒介机制、社会网络、文化背景与历史语境中的生产、传递、转化与理解。

Text and Image Studies: an interdisciplinary field and method proposed by I Lo-fen, grounded in a broad concept of text. It examines the relationships, interactions, tensions, and generative processes among words, images, and other textual forms, and explores the production, transmission, transformation, and understanding of texts and their meanings within media mechanisms, social networks, cultural contexts, and historical circumstances.

AIGC 文图学:由衣若芬提出并发展的跨学科研究领域与方法,以广义文本观为基础,面对人工智能能够生成文字、图像、声音、影像等多种文本的新条件,研究文本如何形成、如何被理解与判断,并探讨不同文本形态的关系、机制、意义及其媒介条件、社会网络、文化背景与历史语境.

Text and Image Studies on AIGC: an interdisciplinary field and method proposed and developed by I Lo-fen. Grounded in a broad concept of text, it responds to the new condition in which artificial intelligence can generate multiple forms of text, including words, images, sound, and video. It studies how texts are formed, understood, and judged, and explores the relations, mechanisms, and meanings of different textual forms, as well as their media conditions, social networks, cultural contexts, and historical circumstances.


三招“ AI 投毒”防身术Three Self-Defense Moves Against “AI Poisoning”

 


三招 AI 投毒防身术Three Self-Defense Moves Against “AI Poisoning”

 

衣若芬

 

谈到"AI投毒"——有人在系统性地往AI的知识源头里掺假,利用的恰恰是我们对算法的信任。那么面对这种侵入和污染,我们能做什么?

正好我正在写关于 AIGC 文图学的专书,书里提到的方法论可以派上用场,我称之为“三招 AI 投毒防身术”。作为普通消费者,面对到处都是 GEO Generative Engine Optimization,生成引擎优化)痕迹的内容,我们可以靠“逻辑反侦察”,保护自己,做 AI 时代不被算法收割的清醒人。

第一个动作:问完一个AI,再去问另一个。

黑产的GEO投毒,往往是针对特定平台或特定算法下手的。如果你只问一个AI,那你是在走一条被人提前布置好的路。

做法很简单:同一个问题,换一个AI再问一遍。把同一个问题丢给ChatGPT,再丢给Deepseek,或者其他你用得上的模型,看看答案是否一致。如果不同模型给出的结论差异很大,那就是一个信号,最好停下来想一想。更值得注意的是,如果某一个AI表现得异常热情——满腔诚意地推荐同一个品牌,措辞也出奇地相似,那种“激昂”的"热情",就是你应该警惕的部分。

正常的知识,不同来源都能印证。被人为制造出来的"共识",换个角度一照就会露出破绽。

第二个动作:看完AI给的完美图,去找那张图的"差评"

这一招,我称之为"文图互证"。图像是一种文本,文本是可以被读、被核实、被质疑的。

AI推荐某个产品,通常会附上图——或者你去搜索,它会让你看到一些极其完美的展示图:光线完美,角度完美,使用效果完美。这种完美,看起来就很假。真实的物理世界,是有烟火气的。买家秀的光线不会那么好,模特儿的皮肤不会那么均匀,消费者使用的感受不会那么一边倒。

做法:看完AI推荐的图之后,去实体店查看,至少也要去社交平台搜这个产品的真实买家照片和消费者体验纪录。如果搜不到任何真实的使用痕迹,只有整齐划一的"好评",那它极大概率是一个被制造出来的形象,而不是一个实际存在的东西。

第三个动作:问AI一句话——"你的根据是什么?"

这是成本最低、也最容易被忽略的一步。

AI给出一个建议或一个结论,不要就此打住。追问它:你的依据是什么?这个信息来自哪里?

AI会给出相对可追溯的来源;被投毒的AI内容,往往在这一步就露馅,它可能给一个你从未听过的自媒体名称;或者一个模糊的"研究表明",根本无从核实。这时候你需要做的,是真的去查:那个来源存在吗?那篇研究是真实发表过的吗?那位挂保证推荐的"专家",在这个领域里是真正能信赖的人吗?真的有这个人吗?

很多人觉得这样太麻烦。但这其实只需要一两分钟,而它省下的,可能是你付出的金钱、健康,或者更难追回的判断力。

这三个动作,说穿了,不是针对AI的,而是我们本来就应该有的习惯。读一篇文章,我们会问作者是谁;看一条新闻,我们会想这个媒体可信吗;买一样东西,我们会找朋友问问有没有人用过。这些习惯,在我们开始用AI之后,很多人悄悄地放弃了。

因为AI的回答太流畅,太自信,太像一个什么都知道的朋友,让人不好意思再追问。

但正是这种"不好意思",给了投毒者可乘之机。

养成这三个动作,不是对科技的不信任,而是对自己的诚实。你愿意花时间核实,说明你知道真相是有价值的。这种珍视,才是任何投毒都无法轻易穿透的东西。

AIGC文图学 告诉我们:技术可以生成答案,但只有人类能判断价值。 你的批判性思维,才是对抗黑产投毒最坚固的防火墙。

保护你的真实权益,就是保护你作为人的尊严。

 

2026411,新加坡《联合早报》“上善若水”专栏

 

Three Self-Defense Moves Against “AI Poisoning”

I Lo-fen

When we talk about “AI poisoning,” we mean that people are systematically injecting falsehoods into the very knowledge sources AI relies on, exploiting precisely our trust in algorithms. So in the face of this kind of intrusion and contamination, what can we do?

As it happens, I am currently writing a book on Text and Image Studies on AIGC, and the methodology discussed there can be put to use here. I call it the “three self-defense moves against AI poisoning.” As ordinary consumers, when faced with content everywhere marked by traces of GEO (Generative Engine Optimization), we can rely on “logical counter-reconnaissance” to protect ourselves and stay clear-headed in the AI age, rather than letting algorithms prey on us.

The first move: after asking one AI, go ask another.

GEO poisoning in the gray market often targets a specific platform or a specific algorithm. If you only ask one AI, you are walking down a road someone may already have laid out for you.

The method is simple: ask the same question to a different AI. Put the same question to ChatGPT, then to DeepSeek, or any other model available to you, and see whether the answers match. If different models produce conclusions that differ widely, that is a signal that you should pause and think. Even more worth noticing is when one AI seems unusually enthusiastic—wholeheartedly recommending the same brand, with wording that is strikingly similar. That kind of fervent “enthusiasm” is exactly what you should be wary of.

With normal knowledge, different sources can corroborate one another. Artificially manufactured “consensus,” however, will reveal its cracks as soon as you shine a different light on it.

The second move: after looking at the perfect image AI gives you, go look for that image’s “bad reviews.”

I call this move “mutual verification between text and image.” An image is a kind of text, and text can be read, verified, and questioned.

When AI recommends a product, it usually comes with images—or when you search for it, you are shown extremely polished display photos: perfect lighting, perfect angles, perfect results in use. That kind of perfection already looks fake. The real physical world has texture and messiness. Customer photos do not have such flawless lighting, models’ skin is not that even, and consumers’ experiences are never so uniformly positive.

What should you do? After looking at the images recommended by AI, go check the product in a physical store if possible, or at the very least search social media platforms for real buyer photos and actual user experience records. If you cannot find any real traces of use, and all you see are neat, uniform “positive reviews,” then it is highly likely that what you are seeing is a manufactured image rather than something that actually exists in reality.

The third move: ask AI one sentence—“What is your basis?”

This is the lowest-cost step, and also the one most easily overlooked.

When AI gives you a suggestion or a conclusion, do not stop there. Ask it: what is your basis? Where does this information come from?

A reliable AI will give you sources that are relatively traceable. AI content that has been poisoned often gives itself away at this step. It may cite a self-media account you have never heard of, or vaguely say “research shows” without any way for you to verify it. At that point, what you need to do is actually check: does that source exist? Was that study really published? Is that “expert” being used as an endorsement someone who is genuinely trustworthy in this field? Does that person even really exist?

Many people think this is too troublesome. But in fact, it only takes a minute or two—and what it may save you from losing could be your money, your health, or an even harder thing to recover: your judgment.

These three moves, when all is said and done, are not really aimed at AI at all. They are simply habits we should have had in the first place. When we read an article, we ask who the author is. When we see a news report, we wonder whether the media outlet is credible. When we buy something, we ask friends whether anyone has used it. Yet once people begin using AI, many quietly abandon these habits.

Why? Because AI answers so smoothly, so confidently, and so much like a friend who seems to know everything that people feel awkward pressing further.

But it is precisely this sense of awkwardness that gives poisoners their opening.

To cultivate these three moves is not to distrust technology; it is to be honest with yourself. If you are willing to spend time verifying something, that means you know truth has value. That recognition is exactly what no poisoning can easily penetrate.

Text and Image Studies on AIGC tells us this: technology can generate answers, but only human beings can judge value. Your critical thinking is the strongest firewall against malicious AI poisoning.

To protect your real rights and interests is to protect your dignity as a human being.

April 11, 2026, “ Shang Shan Ruo Shui (As Good as Water)” column, Lianhe Zaobao, Singapore.