提醒：点这里加小编微信（领取免费资料、获取最新资讯、解决考教师一切疑问！）
Imagine this. You need an image of a balloon for a work presentation and turn to an Al text-to-image generator to create a suitable image. You enter the prompt (提示词)“'red balloon against a blue sky" but the generator returns an image of an egg instead.
What's going on? The generator you're using may have been “poisoned”. What does this mean? Text-to-image generators work by being trained on large datasets that include millions or billions of images. Some of the generators have been trained by indiscriminately (不加选择地) scraping online images, many of which may be under copyright, This has led to many copyright infringement (侵害) cases where artists have accused big tech companies of stealing and profiting from their work.
This is also where the idea of “'poison” comes in. Researchers who want to empowel individual artists have recently created a tool named “Nightshade” to fight back against unauthorised image scraping. The tool works by slightly altering an image's pixels (像索) in a way that confuses the computer vision system but leaves the image unaltered to a human's eyes. If an organization then scrapes one of these images to train a future Al model, its data pool becomes “'poisoned". This can result in mistaken learning, which makes the generator return unintended results. As in our earlier example, a balloon might become an egg.
The higher the number of “poisoned" images in the training data, the greater the impact.Because of how generative Al works, the damage from “"poisoned” images also affects related prompt keywords.
Possibly, tools like Nightshade can be abused by some users to intentionally upload “'poisoned" images in order to confuse Al generators. But the Nightshade's developer hopes the tool will make big tech companies more respectful of copyright. It does challenge a common belief among computer scientists that data found online can be used for any purpose they see fit.
Human rights activists, for example, have been concerned for some time about the indiscriminate use of machine vision in wider society. This concern is particularly serious concerning facial recognition, There is a clear connection between facial recognition cases and data poisoning, as both relate to larger questions around technological governance. It may be better to see data poisoning as an innovative solution to the denial of some fundamental human rights.
8. The underlined word “'scraping” (para. 2)is closest in meaning to
A. Collecting B. enhancing C.damaging D.improving
9. According to the passage, adding poisoned data might
A. increase the exactness of returned information
B. lead users to forget the prompt key words
C. discriminate against great masterpieces
D. cause trouble to the training of generative Al
10. What can be inferred from the last two paragraphs?
A. Computer scientists has learned to respect the copyright of most artists.
B. Data poisoning is somehow reasonable to direct attention to human rights
C. The issue oftechnological governance has aroused the lawyers’ interest
D. Nightshade is being improperly used by human rights activists to recognize faces
11. Which of the following might be the best title of the passage?
A. Data Poisoning: Limiting Innovation or Empowering Artists.
B. Data Poisoning: Risks and Rewards ofGenerative Al Data Training
C. Data Poisoning: Addressing Facial Recognition Issues Among Artists
D. Data Poisoning: Government Empowering Citizens to Protect Themselves.
提醒：点这里加小编微信（领取免费资料、获取最新资讯、解决考教师一切疑问！）
