Welcome to the AI Daily column! Here, you'll find your daily guide to exploring the world of artificial intelligence. Each day, we bring you the hottest topics in the AI field, focusing on developers to help you understand technological trends and discover innovative AI product applications.
Fresh AI Products Click to Learn More: https://top.aibase.com/
1. AI Audio Company Confirms "Three Sheep Audio Gate" as a Clone Event
Shenzhen Ruiyu Technology Co., Ltd. has confirmed that the audio was cloned using their independently developed AI voice synthesis platform. Enhanced security measures have been implemented. Police have arrested the suspect responsible for spreading fake audio and rumors, with the case under further investigation.
AiBase Highlights:
🔍 Shenzhen Ruiyu Technology Co., Ltd. confirms the audio was cloned using their AI voice synthesis platform and has strengthened security measures.
👮♂️ Police have arrested the suspect, Wang, for spreading fake audio and rumors and have taken criminal enforcement actions, with the case under further investigation.
🔗 The company has initiated internal review mechanisms, enhanced user identity verification, and security controls for audio generation to ensure the legal and compliant use of technology.
2. Meitu's AI Short Film Creation Tool MOKI Fully Opens, Supporting Music Generation and Detailed Edits
Meitu Inc. has announced that its AI short film creation tool, MOKI, is now open to all users, offering a one-stop solution for film production. Combining AI-driven scriptwriting, storyboarding, video generation, voiceover, and music capabilities, MOKI allows ordinary users to easily create personalized short films. MOKI excels in Chinese semantic understanding accuracy and Eastern visual styles, with plans to offer more professional features in the future.
AiBase Highlights:
🎬 MOKI focuses on AI short film creation, providing production for animated shorts, web dramas, storybooks, and music videos.
🤖 MOKI is user-friendly, requiring only three steps to generate creative short films, including setting up, content generation, and post-production.
🎥 MOKI supports one-click generation of video content with music, although the current generation quality needs improvement, the product's capabilities have been further enhanced.
3. Magnific Launches V2 Image Generation Service, Directly Outputting 4K Images
Magnific's recently launched V2 image generation service has stirred new waves in the AI image processing field, marking its transition from high-end software to a top-tier AI image generator, showcasing its ambition for technological innovation and user experience optimization.
AiBase Highlights:
✨ Direct 4K image generation: The V2 version can directly generate 4K quality images, improving efficiency and expanding the creative space for creators.
🎨 Diverse style options: Users only need to input descriptive prompts, select resolution and style, and the system generates optimized images, offering rich creative possibilities.
🔧 Fine control and advanced features: Provides fine-tuning tools such as creative detail generation, background adjustment, relighting, etc., to meet the needs of professional users.
Details link: https://top.aibase.com/tool/magnific-ai
4. OpenAI's New Model o1 Claims to "Perfectly" Correct Bias, But Data Does Not Fully Support
Recently, Anna Makanju, Vice President of Global Affairs at OpenAI, spoke at the United Nations "Future Summit" about AI bias. She mentioned that the o1 model can significantly reduce bias in AI systems, but actual test results have not met expectations.
AiBase Highlights:
🌟 The o1 model is said to significantly reduce AI bias, but test results show its performance is below expectations.
💡 The o1 model outperforms GPT-4o in implicit discrimination but performs worse in explicit discrimination.
💰 The o1 reasoning model is costly and runs slowly, requiring future improvements in multiple aspects.
5. Google NotebookLM Adds New Feature: Quickly Summarize YouTube Highlights
Google NotebookLM has introduced a new feature that analyzes video transcript text to distill key information, allowing users to easily access the essence of YouTube videos and enhance learning efficiency. Users can click on topics for detailed information, ask questions, and receive suggestions, with support for audio recording functions to help users search transcript content and create study guides.
AiBase Highlights:
📝 NotebookLM can analyze YouTube video transcripts to distill key information and themes.
🔍 Users can click on topics for detailed information and ask questions, with NotebookLM also providing question suggestions.
🎧 Newly supported audio recording function allows users to search transcript content and create study guides.
6. Over 100 Companies Including OpenAI, Google Support the EU AI Act
The EU's latest AI initiative has attracted support from over 100 companies, including tech giants like OpenAI and Google. Although Apple and Meta have chosen to wait and see, the industry's focus on and sense of responsibility for AI development remains significant. The initiative aims to promote corporate self-regulation and lay the foundation for future AI development.
AiBase Highlights:
🤖 Corporate self-regulation: The agreement encourages the formulation of AI governance strategies, identification of high-risk systems, and enhancement of employee literacy.
💼 Active participation: Over 100 companies have signed the agreement, spanning tech, healthcare, banking, and other industries.
⚠️ Apple and Meta are观望: Concerns about regulatory environment uncertainty and data breach risks, with Meta halting Llama model services in the EU.
7. Runway Launches a 100-Film Support Program
Runway Inc. has announced a 100-film fund program aimed at supporting the application of AI technology in film production. The program offers $5 million in cash grants to attract directors. The company emphasizes flexibility and support for various video project formats.
AiBase Highlights:
🎥 Runway Inc. has established a $5 million fund to support 100 original film projects.
💡 The 100-film fund program supports video projects using the company's generative video models.
🌟 Runway invites renowned figures to serve as advisors, committed to discovering practical and pioneering applications of AI-generated video.
8. US FTC Takes Strong Action: Five AI Companies Sued for False Advertising!
The Federal Trade Commission (FTC) in the United States has taken legal action against five AI companies涉嫌虚假宣传和欺诈行为, emphasizing the protection of consumers and market fairness. Some companies have reached settlements, but others have chosen to contest the allegations. Experts are concerned that regulatory measures may stifle innovation.
AiBase Highlights:
🛡️ The FTC has filed lawsuits against five AI companies, accusing them of false advertising and fraudulent behavior.
💰 DoNotPay has reached a settlement with the FTC, paying $193,000 and promising to stop misleading promotions.
✍️ Rytr has been accused of providing a function to generate fake consumer reviews and has agreed to stop the related service.
9. Deloitte: 72% of Professionals Rank Data Privacy of Generative AI as Their Top Concern!
A recent report from Deloitte highlights the importance of data privacy in generative AI, with concerns rising among professionals. Safety and security are listed as one of the ethical technology principles by 78% of business leaders, and the implementation of the EU AI Act has a profound impact on companies.
AiBase Highlights:
🔒 Data privacy has become the main concern in 2024 for generative AI, with attention rising from 22% to 72%.
📈 78% of business leaders list "safety and security" as one of the ethical technology principles, emphasizing the importance of safety.
⚖️ The implementation of the EU AI Act has a profound impact, prompting companies to make necessary adjustments and compliance in AI use.
10. Baidu World Conference 2024 is About to Begin
Baidu has announced that the 2024 World Conference will be held in Shanghai, showcasing the latest technology and products. The daily call volume of the Wenxin large model 4.0 exceeds 700 million times, with an astonishing growth rate. Baidu has successively launched the tool version and Turbo version to enhance user experience. The Baige AI heterogeneous computing platform 4.0 and the Qianfan large model platform 3.0 have been upgraded to help enterprises use large models more efficiently. The intelligent ecosystem is rapidly expanding, with a large number of merchants developing intelligent agents on the platform.
AiBase Highlights:
🚀 The daily call volume of the Wenxin large model 4.0 exceeds 700 million times, with growth exceeding 13 times.
💡 Baidu has successively released the tool version and Turbo version to enhance user experience.
💼 The Baige AI heterogeneous computing platform 4.0 and the Qianfan large model platform 3.0 have been upgraded to help enterprises use large models more efficiently.
11. OpenAI Launches New Multimodal Content Moderation Model: Based on GPT-4o, Capable of Detecting Text and Images
The digital era places increasing emphasis on content security. OpenAI has introduced a multimodal content moderation model called "omni-moderation-latest," based on the latest GPT-4o technology, capable of accurately identifying harmful text and images. This update provides developers with powerful tools to build stronger moderation systems. The new model supports moderation of text and image inputs, especially excelling in processing non-English content, improving accuracy and detection capabilities, ensuring users communicate in a safer space.
AiBase Highlights:
📊 The new model is based on GPT-4o technology, supporting multimodal moderation of text and images.
🌍 Detection accuracy has improved by 42% in 40 languages, especially in low-resource languages.
🔒 Added two types of text moderation, enhancing the ability to identify illegal and violent content.
Details link: https://openai.com/index/upgrading-the-moderation-api-with-our-new-multimodal-moderation-model/
