内容来源：https://www.technologyreview.com/2025/10/17/1125193/ai-art-artist-new-chapter/

内容总结：

人工智能艺术正从争议漩涡走向顶级画廊。尽管生成式AI作品一度被戏称为"数字垃圾"，但一批先行者正突破工具局限，在艺术领域开辟新天地。

随着Midjourney、Runway等工具降低创作门槛，社交媒体涌现出海量AI图像。但真正值得关注的是，部分创作者正将这些工具纳入艺术实践。音乐人雅各布·阿德勒凭借AI生成影片斩获电影节大奖，他认为"AI只是创作者工具箱的新成员"。

这种易用性带来双重效应。艺术家亨利·多布雷兹指出，工具解放了曾被技术门槛阻挡的创作群体。但若要创作杰作，仍需艺术敏感度与审美判断力，"我们将进入由品味驱动的新时代"。

对科班出身的艺术家而言，AI不仅是捷径。贝丝·弗雷最初痴迷于AI生成的怪异图像，当模型日趋完美后，她反而因"故障感消失"而失去兴趣。而将作品纳入丹佛美术馆永久收藏的基拉·克索里卡则认为，AI的不可预测性恰恰能拓展创作边界。

尽管AI艺术已进入苏富比拍卖行和美术馆体系，但公众接受度仍存鸿沟。在多布雷兹看来，只要社会尚未普遍认同AI作为创作工具的属性，探索之路就始终伴随着"酸甜交织的复杂滋味"。

（注：为保证中文读者阅读流畅，部分外国人名采用通用译法或保留原名，专业术语如"Midjourney"等已作必要解释性处理）

中文翻译：

从粗制滥造到登陆苏富比？AI艺术迈入新阶段

与许多新兴艺术运动相似，生成式AI艺术始终伴随着广泛争议。但仍有艺术家在不断突破这些新工具的创作边界。

在这个AI内容泛滥的时代，声称Midjourney、Runway等工具能创作艺术似乎荒诞不经——诸如"虾仁耶稣""芭蕾卡布奇诺"这类作品有何艺术价值？然而在这片混沌中，确实有人正以审慎态度运用AI工具进行创作。部分AI艺术家已崭露头角：他们在网络收获大量拥趸，作品亮相拍卖行，甚至进驻美术馆展厅。

"有时需要相机，有时需要AI，有时则需要颜料铅笔或其他媒介。"音乐家雅各布·阿德勒凭借作品《全像素空间》在Runway第三届年度AI电影节斩获大奖，"这只是创作者工具箱里新增的一件工具。"

生成式AI最显著特质在于其低门槛。无需专业训练，人们就能在顷刻间将想象化为任意风格的图像。这也正是其备受诟病的主因：Instagram和TikTok正被空洞无聊的内容淹没，企业也不再需要雇佣专业艺术家来生成图像视频。

艺术家亨利·多布雷兹为比特币NFT创作的视觉作品在苏富比拍出2.4万美元，他现任谷歌首位驻场电影制作人。在他看来，这种普惠性正是生成式AI的积极特质："许多放弃艺术表达或无缘钻研技艺的人，如今都能创作分享作品。"

但这不意味着AI杰作会轻易诞生。"我不认为生成式AI能造就天才辈出的时代，"自称"AI辅助艺术家"的多布雷兹表示，"虽然DALL-E等工具无需技术功底，但要创造出有趣内容并准确评判，仍需想象力与艺术敏感度。我们将进入由审美驱动的新时代。"

对于专业艺术家而言，AI的价值远不止捷径。科班出身的贝丝·弗雷在Instagram拥有逾十万粉丝，她最初被AI作品的诡异特质吸引——那些变形的手部与令人难忘的进食画面令她着迷。随着模型缺陷被逐步修正，她已逾年未在平台发布AI作品："越完美越乏味，现在需要费尽心思才能制造故障感。"

AI创作往往意味着控制权的让渡——既受制于工具开发商的技术更新，也受限于工具本身。自称"AI协作艺术家"的基拉·克索里卡认为，这种不可控性正是魅力所在。其短片《诡术师》作为首件生成式AI作品被丹佛美术馆永久收藏："AI最吸引我的是不可预测性。若能接纳这点，它就能拓展创作思路。"

然而无论将AI视作合作者还是艺术媒介，这种观念远未获得普遍认同。对多数人而言，"AI艺术"仍与"AI垃圾"画等号。多布雷兹虽感激已获认可，但面对重重阻力开拓新艺术形式，其心境复杂难言："只要大众尚未接受AI与其他工具别无二致——有人能创造杰作，也有人产出糟粕——这种酸甜交织的滋味就会持续。"

深度聚焦
人工智能

· 与AI聊天机器人建立情感联结竟如此简单：越来越多人对聊天机器人产生依赖。这对部分人无伤大雅，对另一些人则隐患重重。

· 谷歌首次披露单次AI指令能耗数据：这家科技巨头迄今最透明的能耗评估，为研究者提供了期待已久的数据窗口。

· 心理咨询师秘密使用ChatGPT引发客户抵触：某些治疗师在疗程中使用AI，此举正在危及客户的信任与隐私。

· 鸽子为何是AI突破的幕后功臣：这种鸟类从未因智慧获得赞誉，但推动尖端AI系统发展的强化学习技术，其思维模式更接近鸽子而非人类。

保持联系
欢迎订阅《麻省理工科技评论》
获取独家优惠、前沿资讯、活动预告精选内容

英文来源：

From slop to Sotheby’s? AI art enters a new phase
Like many nascent artistic movements, generative AI art has been widely criticized. But some artists are nevertheless pushing the creative limits of these new tools.
In this era of AI slop, the idea that generative AI tools like Midjourney and Runway could be used to make art can seem absurd: What possible artistic value is there to be found in the likes of Shrimp Jesus and Ballerina Cappuccina? But amid all the muck, there are people using AI tools with real consideration and intent. Some of them are finding notable success as AI artists: They are gaining huge online followings, selling their work at auction, and even having it exhibited in galleries and museums.
“Sometimes you need a camera, sometimes AI, and sometimes paint or pencil or any other medium,” says Jacob Adler, a musician and composer who won the top prize at the generative video company Runway’s third annual AI Film Festival for his work Total Pixel Space. “It’s just one tool that is added to the creator’s toolbox.”
One of the most conspicuous features of generative AI tools is their accessibility. With no training and in very little time, you can create an image of whatever you can imagine in whatever style you desire. That’s a key reason AI art has attracted so much criticism: It’s now trivially easy to clog sites like Instagram and TikTok with vapid nonsense, and companies can generate images and video themselves instead of hiring trained artists.
Henry Daubrez, an artist and designer who created the AI-generated visuals for a bitcoin NFT that sold for $24,000 at Sotheby’s and is now Google’s first filmmaker in residence, sees that accessibility as one of generative AI’s most positive attributes. People who had long since given up on creative expression, or who simply never had the time to master a medium, are now creating and sharing art, he says.
But that doesn’t mean the first AI-generated masterpiece could come from just anyone. “I don’t think [generative AI] is going to create an entire generation of geniuses,” says Daubrez, who has described himself as an “AI-assisted artist.” Prompting tools like DALL-E and Midjourney might not require technical finesse, but getting those tools to create something interesting, and then evaluating whether the results are any good, takes both imagination and artistic sensibility, he says: “I think we’re getting into a new generation which is going to be driven by taste.”
Even for artists who do have experience with other media, AI can be more than just a shortcut. Beth Frey, a trained fine artist who shares her AI art on an Instagram account with over 100,000 followers, was drawn to early generative AI tools because of the uncanniness of their creations—she relished the deformed hands and haunting depictions of eating. Over time, the models’ errors have been ironed out, which is part of the reason she hasn’t posted an AI-generated piece on Instagram in over a year. “The better it gets, the less interesting it is for me,” she says. “You have to work harder to get the glitch now.”
Making art with AI can require relinquishing control—to the companies that update the tools, and to the tools themselves. For Kira Xonorika, a self-described “AI-collaborative artist” whose short film Trickster is the first generative AI piece in the Denver Art Museum’s permanent collection, that lack of control is part of the appeal. “[What] I really like about AI is the element of unpredictability,” says Xonorika, whose work explores themes such as indigeneity and nonhuman intelligence. “If you’re open to that, it really enhances and expands ideas that you might have.”
But the idea of AI as a co-creator—or even simply as an artistic medium—is still a long way from widespread acceptance. To many people, “AI art” and “AI slop” remain synonymous. And so, as grateful as Daubrez is for the recognition he has received so far, he’s found that pioneering a new form of art in the face of such strong opposition is an emotional mixed bag. “As long as it’s not really accepted that AI is just a tool like any other tool and people will do whatever they want with it—and some of it might be great, some might not be—it’s still going to be sweet [and] sour,” he says.
Deep Dive
Artificial intelligence
It’s surprisingly easy to stumble into a relationship with an AI chatbot
We’re increasingly developing bonds with chatbots. While that’s safe for some, it’s dangerous for others.
In a first, Google has released data on how much energy an AI prompt uses
It’s the most transparent estimate yet from one of the big AI companies, and a long-awaited peek behind the curtain for researchers.
Therapists are secretly using ChatGPT. Clients are triggered.
Some therapists are using AI during therapy sessions. They’re risking their clients’ trust and privacy in the process.
Why we should thank pigeons for our AI breakthroughs
The bird has never gotten much credit for being intelligent. But the reinforcement learning powering the world’s most advanced AI systems is far more pigeon than human.
Stay connected
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.