夜雨聆风【Artificial Intelligence】
How to spot an AI-generated face, according to science
Training people to pay attention to the right visual cues nearly doubled how accurately they could spot AI-generated faces
SEBASTIAN KAULITZKI/SCIENCE PHOTO LIBRARY via Getty Images
It used to be easy to tell when a face was generated with artificial intelligence (AI). Whether it was a distinctive uncanny sheen, impossibly smooth skin, eyes that didn’t quite make sense or a conspicuous third ear, older AI models’ facsimiles of human faces were simple to spot and easy to dismiss. That’s just not true anymore.
Now, AI image generators can produce portraits so convincing that even careful observers struggle to distinguish fact from figment. That’s why apps such as Zoom and Tinder allow their users to submit biometric identification, such as retinal scans, to help prove that a real person exists behind a profile picture. But a new study suggests you can train your brain to get better at spotting fakes.
Past attempts to teach people to spot AI faces have focused on training viewers to look for visual glitches or statistical fingerprints left behind by a particular image generator, such as a wonky ear or an eye with two pupils. The problem is that those clues can disappear with a software update or by simply using a different prompt. “The AI is getting too good,” said Amy Dawel, an associate professor at Australian National University and the lead author on the study, in a press release. “And fraudsters may avoid using pictures with obvious flaws anyway.” The result is an endless technological arms race humanity seems destined to lose.
Instead, the researchers taught the participants how to recognize broader patterns in how AI systems generate images. “Our training directs people’s attention to global qualities that differ between AI and human faces,” Dawel said.
Current AI image generators are themselves trained on datasets composed of millions of images. When prompted to generate a face, they don’t copy specific faces, but instead compose a new face that is based in part on the mathematical patterns shared across the faces in that data set—these allow the AI to construct a “typical” human face.
The result is that AI-generated faces often drift toward statistical averages. They’re not overly unrealistic, so much as a little too balanced, a little too generic, and a little too conventional. Individually, none of these traits are necessarily suspicious. But together, the whole is blander than the sum of its parts—a subtle banality humans can often implicitly sense.
“Even relatively short training sessions helped participants improve their accuracy,” says Tanya George, a student researcher at Australian National University who trained the study’s participants. “Research like this can help people navigate increasingly complex online environments.”
Compared with real faces, AI-generated faces tend to be more symmetrical, more proportional, and more attractive—while also being less expressive, less distinctive and significantly less memorable. When the researchers trained participants to look for these six markers instead of fleeting artifacts like malformed ears or mismatched jewelry, their ability to spot the AI face almost doubled.
In other words, AI gravitates to the center. Real people do not. Our faces are shaped by countless small deviations from the norm—our subtle asymmetries, distinctive features, and expressions make us memorable. Those imperfections are not flaws. They are our signature.
如何识别AI生成的面孔？
科学给出答案
训练人们关注正确的视觉线索，使他们识别AI面孔的准确率几乎翻了一番。
过去，我们很容易就能分辨出一张脸是不是人工智能生成的。无论是那标志性的诡异油光、不可思议的光滑皮肤、不太对劲的眼睛，还是显眼的第三只耳朵，旧版AI模型仿制的人脸很容易被识破，也容易被人不屑一顾。但现在情况已经不同了。
如今，AI图像生成器能制作出极其逼真的肖像，即便是细心的观察者也难以区分真实与虚构。正因如此，Zoom和Tinder等应用允许用户提交视网膜扫描等生物识别信息，以帮助证明头像背后是一个真实的人。但一项新研究表明，你可以训练自己的大脑，让它更擅长识别假脸。
过去教人们识别AI面孔的尝试，主要侧重于训练观看者寻找特定图像生成器留下的视觉漏洞或统计指纹，比如一只形状怪异的耳朵，或有两个瞳孔的眼睛。问题在于，这些线索可能随一次软件更新，或仅仅换一个提示词就消失了。“AI变得越来越强了，”该研究的第一作者、澳大利亚国立大学副教授艾米·达维尔在一份新闻稿中表示，“而且，诈骗者反正也会避免使用有明显瑕疵的图片。”其结果就是一场人类似乎注定要输掉的无休止的技术军备竞赛。
于是，研究人员转而教导参与者识别AI生成图像时更宏观的模式。“我们的训练将人们的注意力引向AI面孔与真人面孔之间的整体性差异，”达维尔说。
当前的AI图像生成器本身是用包含数百万张图像的数据集训练的。当被要求生成一张面孔时，它们并不会复制某张特定的脸，而是基于该数据集中所有面孔共有的数学模式，组合出一张新面孔——这让AI得以构建出一张“典型”的人脸。
其结果是，AI生成的面孔常常趋向于统计平均值。它们并非过于不真实，而是有点太均衡、太普通、太常规了。单独看，这些特征未必可疑。但合在一起，整体效果比各部分的总和更平淡无味——这是一种人们潜意识里往往能感知到的微妙平庸感。
“即便只是相对简短的培训，也帮助参与者提高了准确率，”负责培训参与者的澳大利亚国立大学学生研究员坦尼娅·乔治说，“此类研究有助于人们驾驭日益复杂的网络环境。”
与真实面孔相比，AI生成的面孔往往更对称、比例更协调、更具吸引力——同时表情更少、辨识度更低，而且远没那么令人难忘。当研究人员训练参与者寻找这六个标志，而非诸如畸形的耳朵或不匹配的耳环这类转瞬即逝的瑕疵时，他们识别AI面孔的能力几乎翻了一番。
换句话说，AI倾向于趋向中心均值。而真实的人不会。我们的面孔由无数微小的偏差塑造而成——那些微妙的不对称、独特的五官和表情，让我们令人难忘。那些不完美之处并非缺陷，而是我们的独特印记。
本文编译自《科学美国人》，原作者Sam Macdonald
https://www.scientificamerican.com/article/how-to-spot-an-ai-generated-face-according-to-science/
这篇关于如何识别AI面孔的文章，最触动我的，不是“准确率翻倍”的训练技巧，而是它揭示的一个略显讽刺的真相：AI追求的那种“完美”，反而成了它最大的破绽。
我们总以为，越逼真越难以分辨，但科学却告诉我们，AI的破绽恰恰在于它太“平均”、太“完美”了。对称的五官、协调的比例、毫无瑕疵的皮肤，这些人类梦寐以求的理想特征，组合在一起竟产生了一种“平淡的平庸感”。这种完美，因为缺少了生活真实的痕迹，反而显得苍白，甚至有点诡异。就像一个永远保持标准微笑、从不犯错的人，你会觉得他亲切，却很难觉得他真实。
更让我觉得温暖的是，文章最后将那些我们平时想要隐藏的“不完美”——比如细微的不对称、独特的表情纹路、不那么规整的五官——定义为“我们的独特印记”。这让我重新审视所谓的“缺陷”。原来，正是那些熬夜留下的黑眼圈、大笑时眼角的细纹、遗传来的小雀斑，甚至是某次意外留下的小伤疤，共同构成了我们独一无二的生命地图。AI可以生成一张完美无瑕的脸，但它永远无法生成一个你在深夜痛哭过、在阳光下大笑过、在岁月里慢慢变老的故事。
这项研究其实也在提醒我们，在一个眼见不再为实的时代，抵抗虚假最好的方式，或许不是去追逐更高级的辨伪技术，而是学会去珍视和感知那些无法被算法量化的真实。真实是有温度的，带着些许笨拙，甚至有点粗粝。就像一棵自然生长的树，它的美，不在于每一片叶子都长在黄金分割点上，而在于它在风雨中形成的、独一无二的姿态。AI永远趋向于那个抽象的中心均值，而鲜活的人，总是带着迷人的偏差。
