«

《阿凡达》导演詹姆斯·卡梅隆表示,生成式人工智能"令人恐惧"。

qimuai 发布于 阅读:68 一手编译


《阿凡达》导演詹姆斯·卡梅隆表示,生成式人工智能"令人恐惧"。

内容来源:https://techcrunch.com/2025/11/30/avatar-director-james-cameron-says-generative-ai-is-horrifying/

内容总结:

在即将上映的《阿凡达:水火之歌》之际,著名导演詹姆斯·卡梅隆在接受哥伦比亚广播公司专访时,对当前热议的生成式人工智能技术表达了鲜明立场。

尽管卡梅隆执导的《阿凡达》系列始终引领视觉特效技术革新,通过表演捕捉将蓝色纳威人塑造得栩栩如生,但他明确划清了表演捕捉与生成式AI的界限。这位技术先驱强调,表演捕捉实为"对演员表演艺术的礼赞",通过记录演员的实时表演为数字艺术家提供创作蓝本,始终以演员与导演的创作互动为核心。

"当我们深入展现创作过程时,观众会发现这完全是对表演艺术的尊崇。"卡梅隆指出。节目镜头记录着演员们在25万加仑水池中完成水下场景的拍摄实况,印证着真人表演不可替代的价值。

谈及生成式AI时,卡梅隆直言"令人担忧":"只需文本指令就能凭空生成虚拟角色、虚构演员、创造表演,这完全背离了我们的创作理念。"在技术革新的浪潮中,这位电影大师坚守着艺术创作中的人文精神。

中文翻译:

詹姆斯·卡梅隆的电影始终走在视觉特效技术的前沿——尤其是《阿凡达》系列,通过表演捕捉技术赋予蓝色纳威英雄鲜活的生命力。
但这并不意味着卡梅隆青睐生成式人工智能。
在为即将上映的《阿凡达:火与灰》接受 CBS《周日早间》节目采访时,这位导演承认表演捕捉(将演员表演记录为数字艺术家的创作模板)听起来可能与生成式人工智能相似,但他强调二者"本质截然相反"。
"多年来总有人误解:'他们用电脑搞些奇怪的东西,是要取代演员',"卡梅隆表示,"但当你深入了解我们的工作方式,会发现这其实是对演员与导演创作时刻的礼赞。"
CBS节目片段展现了《阿凡达》剧组在25万加仑水箱中拍摄水下场景的过程。
"而生成式人工智能则走向另一个极端,"卡梅隆补充道,"仅凭文字指令就能凭空生成角色、虚构演员、捏造表演。这太可怕了......我们的创作理念与此背道而驰。"

英文来源:

James Cameron’s movies are often at the cutting edge of visual effects technology — especially the “Avatar” films, with their heroic blue Na’vi characters brought to life through performance capture.
But that doesn’t make Cameron a fan of generative AI.
In a CBS Sunday Morning interview tied to the upcoming release of “Avatar: Fire and Ash,” the director acknowledged that performance capture (where an actor’s performance is recorded as a template for digital artists) can sound similar to genAI. But in reality, he said it’s “the opposite.”
“For years, there was this sense that, ‘Oh they’re doing something strange with computers, and they’re replacing actors,’” Cameron said. “When in fact, once you really drill down and you see what we’re doing, it’s a celebration of the actor-director moment.”
Indeed, the CBS segment shows members of the “Avatar” cast performing their underwater scenes in a 250,000-gallon water tank.
“Go to the other end of the spectrum and you’ve got generative AI, where they can make up a character, they can make up an actor, they can make up a performance from scratch with a text prompt,” Cameron added. “No, that’s horrifying … That’s exactly what we’re not doing.”

TechCrunchAI大撞车

文章目录


    扫描二维码,在手机上阅读