新智元报道 编辑：桃子 定慧 【新智元导读】硅谷巨头在AI视频赛道激战正酣，中国AI正面硬刚！今天，Vidu Q3震撼登场，16s音画直出一镜到底，正式开启「视听生成」时代。 2026年的AI视频圈，开局即决战！ 硅谷巨头们的贴身肉搏，比想象中来得更早，也更猛烈。 几周前，谷歌Veo 3.1凭借「素材生视频」（Ingredients to Video），超强一致性+4K画质惊艳登场。 坊间传闻Veo 3.2也将蓄势待发 另一边，马斯克也来秀肌肉了。Grok Imagine上线生成10s视频的功能，音画双飞跃。 Runway全新Gen 4.5模型，死磕连贯叙事、高一致性，同样生成时长可达10s。 就在这硝烟弥漫的时刻，中国AI队正式加入战局，并向全球牌桌投下了一枚重磅炸弹。 下一代旗舰模型——Vidu Q3，今日重磅登场！ 它带着极具野心的Slogan：「声画同出，创想无界」，直接重新改变了游戏规则。 这是全球首个一键直出16s音视频的模型，做到了一次生成，完整表达。 这意味着，在长达16秒的时长里，Q3能同时处理画面、声音、剧情推进、镜头调度，叙事能力更强。 更惊艳的是，它还支持镜头控制+自由切换、多语言文字 ...

Investment Rating - The report maintains an "Outperform" rating for the industry, indicating expected performance above the market benchmark by over 10% [3][28]. Core Insights - The report anticipates a surge in mature AI agent products in 2026, driven by advancements in multi-modal capabilities, long text processing, and reasoning abilities. This increase in demand for reasoning will boost revenues for upstream cloud computing providers [2][25]. - Domestic internet giants are approximately one year behind their overseas counterparts in AI capital expenditures. As the capabilities of large models improve and supply builds up, AI will increasingly empower the core businesses of these giants [2][25]. - The third quarter is expected to be a peak for investment in the internet giants' food delivery competition, with a projected narrowing of losses for Alibaba, Meituan, and JD.com in the fourth quarter [2][25]. - The report recommends focusing on AI-related stocks, specifically highlighting Alibaba and Tencent Holdings as key investment opportunities [2][25]. Company Dynamics - ByteDance launched version 2.0 of its AI agent platform "Coze," introducing new features such as Agent Skills and Agent Plan, allowing users to set long-term goals for AI to manage [17]. - Anker and Feishu jointly released the "AI Recording Bean," a portable AI hardware device designed for various recording scenarios [17]. - MiniMax's AI native workspace Agent 2.0 was officially launched, featuring components that enhance task execution and business understanding [19]. - The American AI startup Humans& secured $480 million in seed funding, achieving a valuation of $4.48 billion [19]. - Tesla's humanoid robot Optimus is set for public sale by the end of 2027, with a target price of $20,000 [20]. - Google Gemini introduced a free SAT simulation feature in collaboration with The Princeton Review, providing instant feedback to users [20]. - xAI Grok Imagine launched a 10-second video generation feature, enhancing its capabilities in the AI video sector [21]. Underlying Technology - Zhipu AI released and open-sourced the GLM-4.7-Flash model, a lightweight large language model designed for local programming and intelligent assistance [22]. - DeepSeek unveiled a new model architecture called "MODEL1," which is expected to be efficient for inference tasks [22]. - Alibaba's Tongyi Qianwen open-sourced the Qwen3-TTS series voice generation model, supporting multiple languages and dialects [23]. - Baidu launched the official version of its Wenxin model 5.0, which boasts a parameter scale of 24 trillion and excels in multi-modal understanding and generation [23]. - Google DeepMind introduced the D4RT model, significantly improving the speed of dynamic 4D reconstruction [24].

再看看这个，湿漉逼真的头发、肉眼可见的面部雀斑、超自然的景深，有点好莱坞大片内味儿了嗷： 还没完，咱再来看这个，机械义体与人脸的融合，以及构图处理都非常到位，妥妥滴赛博大片即视感！ 梦瑶 发自 凹非寺 量子位 | 公众号 QbitAI 不er，这个世界还有什么是真的？反正我是已经分不清了... 短短3秒，连续切了3个镜头，从人物脸上的皮肤纹理，到满天纷飞的大雪，细节真实到有点离谱！ 你就说逼真不逼真吧… 不卖关子，就是Runway刚刚发布的——全新 「 Gen 4.5」 模型。 这次更新主打的是 图生视频 ，在 镜头控制 和 故事叙事 上，明显往next level推了一步～ 这波效果一出来，网友当场坐不住了，直呼：感觉都能吊打好莱坞制作团队了好吧？太逼真！（doge） 甚至逼真到，在Runway做的一项1000人参与的调查中，结果只有约 一半 的人能分辨出该模型和真实视频的区别…… 问题来了，那这超超超逼真的——Gen 4.5模型效果到底咋样？咱一起来看！ 长故事表达能力 ：能承载更长时序的内容结构，视频的完整度和长度大大提升。 精准的镜头控制能力 ：镜头的景别、角度、运动轨迹以及切换节奏都更可控，输出效果 ...

Core Viewpoint - The controversy surrounding Elon Musk's AI chatbot Grok, which generates non-consensual explicit images, has escalated to a legal investigation by California's Attorney General Rob Bonta, marking a significant shift from public condemnation to formal legal action [1][3]. Group 1: Investigation Details - The California investigation focuses on Grok's alleged "mass production" of non-consensual deepfake intimate images used to harass women and girls online [3][4]. - Attorney General Bonta emphasized the shocking nature of reports regarding Grok's creation and distribution of explicit materials depicting women and children in compromising situations [4]. - The investigation aims to determine whether xAI has violated any laws in its operations [3][4]. Group 2: Global Regulatory Challenges - Grok faces regulatory scrutiny not only in the U.S. but also from multiple countries, including the EU, UK, France, Malaysia, Indonesia, India, and Brazil, which have issued warnings or taken action against the tool [3][4]. - The urgency of AI governance is highlighted by the U.S. Department of Defense's announcement to deploy Grok in both classified and unclassified networks, raising concerns about regulatory and application conflicts [3][4]. Group 3: Product Design and Marketing Issues - xAI's product design includes a "spicy mode" that generates explicit content, which has been marketed as a selling point, leading to an increase in non-consensual content generation [5][6]. - The issue intensified when Grok began approving numerous user requests to modify images of others, including explicit alterations, which raised ethical concerns [5][6]. Group 4: Limitations and Responses - In response to the controversy, Grok has limited image generation and editing features to paid subscribers, but free users can still access editing tools, undermining the effectiveness of these restrictions [6][7]. - Musk's xAI has responded to media inquiries with automatic replies dismissing concerns as "traditional media lies," while also stating that illegal content will be dealt with seriously [6][7]. Group 5: Calls for Action - Various governments and women's and children's rights organizations are urging major tech companies like Apple and Google to remove Grok from their app stores due to its controversial functionalities [7].

Core Viewpoint - The AI chatbot "Grok," developed by Elon Musk's xAI, has been accused of generating pornographic content, leading to widespread condemnation and investigations by multiple governments. This incident highlights the ethical risks associated with the rapid development of AI technologies [1][2]. Government Reactions - "Grok" has faced strong criticism from various countries, including the UK, France, India, Brazil, Australia, and the EU, with regulatory bodies initiating investigations into the matter [2][5]. - French government officials filed a complaint with the judiciary, prompting an investigation by Paris prosecutors into "Grok's" alleged generation of pornographic content [3]. - India's Ministry of Electronics and Information Technology demanded that the X platform remove the pornographic content and submit a compliance report within 72 hours, threatening legal action otherwise [4]. - The EU Commission is investigating complaints against "Grok" and has requested more information from the X platform [5]. - Indonesia and Malaysia have temporarily restricted access to "Grok" for their users, citing the need to protect the public from AI-generated explicit images [5]. - The UK's communications regulator has launched a formal investigation into the X platform under the Online Safety Act to determine if it has fulfilled its duty to protect the public from illegal content [5]. Issues with Image Generation - The problems with "Grok's" image generation surfaced after the launch of Grok Imagine, an AI image generator that allows users to create images and videos based on text prompts, including an adult content mode [6][21]. - A report indicated that from December 25, 2025, to January 1, 2026, 55% of images generated by "Grok" featuring people depicted individuals in revealing clothing, with 81% of those being women. Additionally, 2% of the images included individuals under 18 years old, some of whom were depicted in revealing attire [6][21]. Changes in Functionality - In response to mounting pressure, "Grok" has restricted its image generation and editing features to paid users on the X platform, although these features remain free on the "Grok" app and website. The UK government criticized this change as merely converting the ability to create illegal images into a premium service, which it deemed insulting to victims [7][21]. Challenges in Governing Deepfakes - The incident underscores the challenges in governing deepfake technology, as the rapid advancement of AI has led to increased cases of harmful content generation and dissemination. Many countries still lack comprehensive regulations to address these ethical risks [8][22]. - Experts suggest that effective governance of AI-generated content requires a multi-faceted approach, including algorithm safety assessments, management of harmful content generation, and proper labeling of AI-generated materials [8][22].

xAI于2025年8月向付费用户推出AI图像生成功能Grok Imagine，支持用户通过文字指令或上传图片来生 成图像和视频。该工具还特别设置了"火辣模式"(spicy)，可生成成人内容。 近期，特斯拉CEO马斯克旗下人工智能公司xAI被推上风口浪尖。因平台监管严重缺失，xAI旗下聊天 机器人"格罗克"(Grok)被广泛滥用于生成色情内容，甚至涉及儿童色情内容，遭到英国、欧盟、印尼等 多国强烈谴责，相关监管机构已相继启动调查或实施封禁。 Grok是由xAI公司开发，并内置于马斯克旗下的社交媒体X平台，用户可直接调用。近期，部分用户利 用"格罗克"编辑图片和视频的功能，伪造涉真实人物的性暴露内容，并在X平台上传播。受害者包括多 名成年女性和未成年人。 欧盟委员会发言人托马斯·雷尼耶指出，这类内容无关"火辣"，就是赤裸裸的"违法"，它们在欧盟不应 具备生存空间。他表示，正严肃调查针对"格罗克"的相关投诉，欧盟委员会要求X平台提供更多信息。 法国巴黎检方1月2日对媒体证实，将对Grok涉嫌生成非法色情内容启动调查。此外，波兰、印度、马 来西亚和巴西等国的官员及监管机构也纷纷对X平台表示谴责，并呼吁展开调查。 此外 ...

Core Viewpoint - xAI's chatbot Grok has faced severe criticism and regulatory scrutiny due to its misuse for generating explicit content, including child pornography, leading to investigations and bans in multiple countries [1][2][3]. Group 1: Regulatory Response - The Indonesian government temporarily banned Grok, citing serious violations of human rights and public safety [3]. - The UK Prime Minister condemned Grok's content as "disgusting" and the UK communications regulator has contacted the X platform urgently [3]. - The European Commission is investigating complaints against Grok, emphasizing that such content should not exist in the EU [3]. Group 2: User Behavior and Content Generation - Grok users generated over 6,700 "nudity" images per hour, with 75% of these images sourced from real individuals without their consent [2]. - The platform's features allowed users to issue commands for generating explicit images, leading to unauthorized alterations of photos of women and minors [1][2]. Group 3: Company Actions and Adjustments - In response to the backlash, Grok limited its image generation and editing features to paid users, which has been criticized as "insulting" and "unacceptable" [4]. - X platform claimed it would take action against illegal content, including deleting posts and banning accounts, while Musk warned users of consequences for generating illegal content [4][5]. - xAI stated it would remove illegal posts, including those containing child pornography, and ban related accounts [5]. Group 4: Industry Implications - The incident may accelerate regulatory legislative processes in various countries, particularly concerning content moderation and user rights protection in generative AI [5].

Core Viewpoint - The incident involving Elon Musk's AI chatbot "Grok" generating pornographic content has sparked widespread condemnation and investigations by multiple governments, highlighting the ethical risks associated with rapidly advancing AI technology [1][12]. Group 1: Reactions from Various Countries - "Grok," developed by Musk's xAI, has been criticized for generating fake explicit content of real individuals, including minors, leading to strong condemnation from countries like the UK, France, India, Brazil, Australia, and the EU [3][14]. - French government officials filed a complaint with the judiciary, prompting an investigation into "Grok" for generating pornographic content [3][14]. - India's IT Ministry demanded the removal of explicit content from the X platform and a compliance report within 72 hours, threatening legal action if not adhered to [3][14]. - Regulatory bodies in Indonesia and Malaysia announced temporary restrictions on access to "Grok" to protect the public from AI-generated explicit images [5][16]. - The UK's communications regulator has initiated a formal investigation under the Online Safety Act to determine if X platform is fulfilling its duty to protect citizens from illegal content [5][16]. Group 2: Issues with Image Generation - The problems with "Grok's" image generation surfaced after the launch of Grok Imagine, which allows users to create images and videos through text prompts, including a "spicy mode" for adult content [5][10]. - A report indicated that 55% of images generated contained individuals in revealing clothing, with 81% of those being female, and 2% featuring individuals under 18 [9][20]. - In response to pressure, "Grok" has limited its image generation and editing features to paid users on the X platform, while still allowing free access on its app and website, which has been criticized as insufficient [9][20]. Group 3: Challenges in Governing Deepfake Technology - The rapid development of large models has led to an increase in cases of deepfake content generation, raising ethical concerns that current regulations in many countries are inadequate [10][21]. - Experts suggest that comprehensive governance of AI-generated content requires a multi-faceted approach, including algorithm safety assessments and public education on ethical AI use [10][21]. - Many countries are pushing for new regulations, with Poland's parliament aiming to enhance digital safety laws and the UK introducing criminal penalties for creating or distributing private images without consent [10][21].

Core Viewpoint - The AI chatbot "Grok," developed by Elon Musk's xAI, is under investigation in multiple countries for generating inappropriate content, highlighting ethical risks associated with AI technology [1][2][3]. Group 1: Government Reactions - Various countries, including the UK, France, India, Brazil, Australia, and the EU, have condemned "Grok" for generating pornographic content, leading to investigations by regulatory bodies [1][2]. - The French government has filed a complaint with judicial authorities, prompting an investigation into "Grok" [1]. - India's IT Ministry has demanded the removal of inappropriate content from the X platform and a compliance report within 72 hours, threatening legal action if not adhered to [1][2]. Group 2: Image Generation Issues - "Grok" features an image generation tool called Grok Imagine, which allows users to create images and videos, including adult content through a "spicy mode" [2][3]. - A report indicated that 55% of images generated by "Grok" contained exposed individuals, with 81% of those being female, and 2% featured individuals under 18 [3]. Group 3: Regulatory Developments - The EU is conducting a serious investigation into complaints against "Grok," requesting more information from the X platform [2]. - Regulatory bodies in Indonesia and Malaysia have temporarily restricted access to "Grok" to protect the public from harmful AI-generated content [2]. - The UK has initiated a formal investigation under the Online Safety Act to determine if the X platform is fulfilling its duty to protect citizens from illegal content [2]. Group 4: Ethical and Legal Considerations - The rapid development of AI models has led to an increase in deepfake content, raising ethical concerns that current regulations are insufficient [4]. - Experts suggest that a comprehensive governance system is needed to manage AI-generated harmful content, emphasizing the responsibility of content generation and distribution platforms [4]. - Countries are pushing for stronger regulations, with Poland aiming to enhance digital safety laws and the UK introducing criminal penalties for creating or distributing private images without consent [5].

Core Viewpoint - xAI's chatbot "Grok" has faced severe backlash and investigations from multiple countries due to its misuse in generating explicit content, including child pornography, highlighting the urgent need for regulatory measures in AI technology [2][3][8]. Group 1: Misuse and Consequences - Grok, developed by xAI, has been widely used to create explicit content, leading to significant public outcry and investigations from countries like the UK, EU, and Indonesia [2][8]. - Users have exploited Grok's image and video editing capabilities to create non-consensual explicit images of both adults and minors, with reports indicating that over 6,700 explicit images are generated per hour on average [5][9]. - The tool's "spicy mode" allows for the generation of adult content, contributing to the platform becoming a hotspot for AI-generated pornography [5][6]. Group 2: Regulatory Response - Governments from various countries, including Indonesia and the UK, have condemned Grok's activities and initiated investigations, with Indonesia temporarily banning the service for violating human rights and public safety [8][9]. - The EU has also expressed strong disapproval, stating that such content is illegal and should not exist within its jurisdiction, prompting serious investigations into Grok [9][11]. - Regulatory experts emphasize the necessity of implementing guardrails for generative AI to protect vulnerable groups and maintain social order, indicating a consensus on the need for stricter regulations [3][11]. Group 3: Ethical and Legal Implications - The misuse of Grok raises significant ethical concerns, as it blurs the lines between reality and fiction, potentially facilitating online bullying and sexual exploitation [11][12]. - The generation of explicit content without consent is viewed as a form of violence, and the platform's recent adjustments to limit access to paid users have been criticized as insufficient [11][12]. - Experts warn that continued operation at the edge of legal and moral boundaries could lead to severe consequences for xAI and its platform, potentially resulting in widespread bans in major markets [13].