中国新歌声
Rise of AI performers triggers concerns about human actors_我的网站
A | 记者24日从兰州大学获悉，该校物理科学与技术学院田永辉教授团队在光学数字计算领域取得重要进展。该团队研制出一种可重构光学算术逻辑单元芯片，将多种逻辑与算术运算集成于同一硬件平台。相关成果发表于国际权威期刊《自然·通讯》。
Fang Taozi, an AI-generated character who stars in the hit AI short drama series The Girl Who Was Fired, has posted lifestyle vlogs, launched social media accounts, shared her breakfasts, workouts and pets, landed brand deals and has a fanbase.
Since the drama debuted on short video platform Douyin on June 12, it has accumulated 200 million views. Fang's Douyin account has attracted more than 400,000 followers.
Fans interact with her as if she were a real celebrity, creating fan edits, discussing fictional relationships and building a community around her imagined life.
This raises a question that has become increasingly urgent for the entertainment industry: Does Fang Taozi represent the first generation of AI celebrities? The answer depends on how "celebrity" is defined, said Zhang Peng, a cultural researcher and associate professor at Nanjing Normal University, in an interview.
If celebrity status is measured by public attention, emotional connection and commercial value, Fang appears to meet many of the criteria. She has fans, a personal brand and even advertising opportunities.
Whether AI can truly become stars depends not on the technology itself, but on their ability to forge genuine emotional connections with audiences and sustain their appeal over time, added Zhang.
However, Shi Wenxue, a Beijing-based cultural critic, argues that AI stars have indeed emerged, but they should be viewed as part of the broader lineage of virtual idols.
From the 1958 US virtual musical group Alvin and the Chipmunks to Hatsune Miku, Luo Tianyi and the hyper-realistic digital human Liu Yexi, virtual performers have existed in various forms for decades.
From the perspective of how she is created, Fang Taozi represents a technological advance. Her emergence is largely a product of the current economic environment of the film and television industry.
The rise of AI performers has also triggered concerns among human actors, especially those working in smaller productions.
Human performers bring personal experience, emotional interpretation and unpredictability - qualities that remain difficult to reproduce. But for industries that require large volumes of affordable content, AI characters offer obvious advantages: They do not age, can work continuously and can be customized for different audiences.
AI short dramas are cheaper, faster and free from the risks associated with human performers, raising concerns over the future of lower-level entertainment jobs in areas such as short dramas, advertising, background acting and voice work, said Zhang.
In Shi's view, beneath the debate lies a familiar tension: technological efficiency versus the threat of displacement for human workers.
The question is not whether AI will completely replace actors, but which parts of acting and entertainment can be automated, he noted.
Rather than presenting an AI figure only as a digital image, Chinese platforms appear to be building complete entertainment ecosystems around AI characters, including dramas, social accounts, fan interactions and commercial partnerships.
China's AI stars have grown out of the audience's desire for emotional companionship, while Hollywood's AI experiments have been shaped by industry battles over efficiency and costs, noted Zhang.
According to Shi, Fang Taozi's sudden popularity is, in many ways, a reflection of audience psychology. Whether they are moviegoers, fans or casual viewers, audiences have grown weary of formulaic content, scripted fan service, predictable performances and fragile celebrity personas. An idol like Fang Taozi offers an alternative: She cannot be involved in scandals, play diva or break character, and her persona can be adjusted to meet audience demands. But the model also comes with risks, particularly the limitations of its creators and the potential for excessive commercialization.
。 随着人工智能与数据密集型应用对高速、低时延计算需求的激增，传统电子集成电路正面临摩尔定律放缓、互连带宽受限和功耗持续上升等挑战。集成光子技术凭借大带宽、高速并行处理和低传输时延等优势，为突破电子计算瓶颈提供了重要的技术路径。
B | 基于这一背景，研究团队依托成熟的硅光子集成平台，利用微环调制器体积小、调制速度快、功耗低且易于电控重构的特点，研制出可实现多种逻辑与算术功能动态切换的光学数字计算芯片。团队通过级联微环构建基本逻辑单元，实现了NOT、AND、OR、XOR和XNOR等多种逻辑运算；同时引入波长维度并行处理不同二进制位，使同一硬件架构能够动态切换为2比特加法器、减法器和数值比较器，无需重新设计芯片结构。
C | 芯片基于220纳米硅绝缘体平台制备，每个微环集成载流子耗尽型高速电光调制结构和热调谐单元，可分别完成高速数据加载与工作波长校准。实验结果表明，该系统在20Gbit/s速率下实现了基本逻辑运算及算术功能，计算密度达到528Gbit/s2，能效达12.36fJ/bit。 在此基础上，团队完成了高速数据加密与解密、医学图像水印嵌入和图像差异增强等应用验证，展示了该架构从基础逻辑器件向多功能光学数字计算系统扩展的能力。 田永辉表示，该工作为构建高密度、低功耗、可重构的光子算术逻辑处理芯片提供了新的实现路径，也为光学数字计算在高速信息处理和片上智能计算中的应用奠定了器件基础。
D | （记者颉满斌 通讯员余若怡）。
Current article:http://www.diaoengngtengdiachanpiezao.sbs/czukn63/12a1.xlsx
Published on:10:59:31