Recently, a new study from the University of Michigan found that an energy-efficient method for training large language models can achieve similar results in the same amount of time while reducing energy consumption by 30%. This method can save enough energy to power 1.1 million American households by 2026. The researchers developed a software tool named Perseus that identifies the critical path, which is a series of subtasks requiring the longest time to complete. Then, Perseus reduces the processor speed on non-critical paths so that all tasks can finish simultaneously.

Microsoft has officially started the AI strategic layout for the Xbox platform. As the first AI feature, an intelligent customer service chatbot has already launched on the Xbox platform. According to informed sources, this is only the first step in Microsoft's push for AI strategy in the Xbox area. The second upcoming AI feature is natural language search. This feature will allow users to obtain recommendations without needing to remember specific game titles; they can simply input descriptive terms like "children's puzzle games". The new search system will also support error tolerance, capable of recognizing typos and abbreviations.

Just a few days after launching the new search feature, ChatGPT experienced service outages on Monday morning, leaving many users frustrated. The outage mainly manifested as the system frequently displaying error messages, especially when users attempted to use the new search feature, which was particularly severe. According to user feedback, ChatGPT's webpage often showed '404 Not Found' error messages, making it difficult to conduct searches smoothly. This issue has garnered widespread attention, with outage reports appearing on DownDetector.

As the U.S. elections approach, AI search engines are facing significant challenges. At this critical moment, major tech companies are striving to provide accurate and reliable information to address concerns about misinformation and the 'hallucinations' generated by AI content. In this competition, Perplexity AI has caught attention with its new AI Election Center. This platform offers real-time updates, incorporating data from the Associated Press and collaborating with the nonprofit voter guidance site Democracy Works.

Recently, AI programming assistants have become incredibly popular, claiming to help programmers write code and improve efficiency. Many programmers even consider it a 'savior', eager to use it for coding every day. However, a study from Stanford University has doused the enthusiasm of these 'zealous fans' with cold water: AI programming assistants may be a 'security nightmare'! Stanford researchers gathered 47 programmers to complete five security-related coding tasks, covering Python, JavaScript, and C languages. The results revealed...

Kunlun Wei Technology Co., Ltd. announced that its self-developed Tian Gong AI advanced search feature has officially released its latest version. This upgrade aims to provide users with a more accurate and professional search experience, especially in the fields of financial investment and academic research. Since its launch, Tian Gong AI search has been dedicated to providing innovative intelligent search and information processing solutions to global users through leading AI technology.

Recently, AI search company Perplexity officially launched its new Election Information Center, aimed at providing voters with reliable voting information. This platform will offer users AI-generated answers to voting-related questions, candidate summaries, and will track real-time voting results on Election Day, November 5, with data sourced from the Associated Press. Perplexity stated that the data provided by this voter information center, including voting requirements, polling locations, and times, is based on the non-profit organization Democra.

Google AI Studio and Gemini API have recently launched the 'Real-Time Connection with Google Search' feature, aimed at helping developers improve the accuracy of AI model responses by integrating real-time data from Google Search. With this update, the Gemini 1.5 model can retrieve the latest information from Google Search, enhancing the accuracy and transparency of responses. Developers can directly access this feature in the 'Tools' section of Google AI Studio or in the Gemini API.

With the development of AI technology, many difficult problems in learning scenarios are being solved efficiently. On November 1, Quark fully upgraded its 'AI Question Search' product, making the process of searching and solving problems faster and more capable. Users can find various new and difficult questions through their smartphones, tablets, and computers, and the AI will provide step-by-step explanations, allowing users to ask questions at any time. After launching the 'One-Stop AI Services' and 'System-Level All-Scenario AI Capabilities', Quark's AI capabilities are being applied to learning scenarios, helping users enhance their learning abilities and efficiency.

Baido's official Weibo announced on October 31 that Baido founder Li Yanhong will deliver a one-hour keynote speech at the Baido World 2024 conference, which will be held on November 12. The poster for this speech integrates key insights from Li Yanhong's recent appearances. On the application level, he focuses on the development directions of natural language processing, intelligent systems, and search technology. Regarding industry trends, Li Yanhong pointed out in a previous interview with the Harvard Business Review that despite the inevitability of bubbles, 1% of companies will still stand out in competition.

Meta recently announced that its MobileLLM model is now open to researchers, allowing users to download and use these models for free on the Hugging Face platform, fully integrated with the Transformers library. This initiative aims to promote research and development of large language models (LLMs) on mobile devices, providing a broader range of tools and resources for developers and academia. MobileLLM represents a significant advancement for Meta in the field of mobile AI.

OpenAI has launched a new feature called ChatGPT search, which allows users to quickly obtain the latest web search results through a conversational interface and provides relevant webpage links. This improvement enables users to receive timely answers directly within ChatGPT without switching to traditional search engines. ChatGPT search combines the advantages of a natural language interface with the value of real-time information, such as sports scores, news, stock quotes, and more.

According to a recent report by The Information, Meta is quietly advancing an ambitious plan to develop a dedicated search engine for its AI chatbots. This strategic shift stems from Meta's deep reflection on the significant losses caused by changes to Apple's privacy policies. The search engine project has been under wraps for at least eight months. The company aims to integrate web indexing into Meta AI, thereby reducing its dependence on Google Search and Microsoft Bing.

Is AI actually starting to ‘get smarter’?! The latest research from MIT indicates that the internal structure of large language models (LLMs) astonishingly resembles that of the human brain! This study utilized sparse autoencoder techniques to conduct an in-depth analysis of the activation space of LLMs, revealing three hierarchical structural features that are truly remarkable: first, at the microscopic level, researchers discovered the existence of structures akin to ‘crystals’. The surfaces of these ‘crystals’ are composed of parallelograms or trapezoids, similar to familiar lexical analogies such as ‘man:’

Recently, some media reports indicated that WeChat is conducting gray testing of the AI Q&A feature, allowing some users to input questions in the WeChat search box to trigger the AI Q&A function. The AI Q&A can generate an approximately 400-word response in about 20 seconds and provide 5 reference materials from WeChat official accounts. The test results show that the AI Q&A feature currently mainly cites WeChat official accounts as reference materials.

Today, OpenAI announced the launch of an eagerly anticipated new feature on the ChatGPT web version: users can now easily search through their chat history. This means that everyone can quickly retrieve previous conversations or continue unfinished chats. The gradual rollout of this feature has excited users, especially those who frequently need to review chat content, as it is incredibly convenient. It is reported that Plus and Team users will gain access to this new feature within a day, while Enterprise and