At the 2024 Bund Conference, Tesla senior instructor Xu Lei shared his insights on the future of autonomous vehicles. Xu believes that once cars achieve full autonomy, they will transition from traditional transportation tools to profit-generating instruments, metaphorically referred to as 'money-printing machines.'

A group of amateur enthusiasts successfully verified that the fifth number BB(5) of the Busy Beaver problem equals 47,176,870 using the Coq proof assistant software. This is the most significant advancement in the field of computational theory since 1983. This achievement marks an enhancement in the ability to solve complex mathematical problems, demonstrating the potential of software-assisted proof. The Busy Beaver problem dates back to 1974 and aims to find the specific Turing machines that can write the most '1's. The core of the problem lies in understanding the behavior of Turing machines, particularly their performance regarding the halting problem. This discovery not only contributes to

At the main forum of the 2024 Inclusion Bund Conference, Michael Jordan emphasized three key points missing from the current discussions on artificial intelligence: collectivity, uncertainty, and incentive mechanisms. He pointed out that AI systems need to form collaborative collectives and introduce an economic perspective on incentives to build collaborative systems. He noted that AI performs inadequately when facing uncertainty and that humanity's ability to cope with uncertainty through collective collaboration is worth emulating. To enhance the collective collaboration capability of AI systems, Michael Jordan suggested constructing a 'three-layer data market' model.

Researchers at Binghamton University in New York, led by Ahmed Abdeen Hammed, have developed a machine learning algorithm called xFakeSci that can identify fabricated academic papers with up to 94% accuracy. Hammed and his team created 50 fake papers on topics including Alzheimer's disease, cancer, and depression, and conducted comparative analyses with genuine papers. They found that the number of bigrams in the fake papers was significantly lower than in the real papers, and the relevance of the content was much stronger. Hammed plans to expand the algorithm to more fields in response to A

Sports technology startup SportAI has completed a $1.8 million seed funding round aimed at leveraging artificial intelligence to enhance athlete performance. Its system focuses on coaching, commentary, and analysis of athletic techniques through machine learning, computer vision, and biometrics, primarily serving coaches, training facilities, media broadcasts, and sports brands. SportAI's cloud platform can analyze athletes' technical performance in real-time, providing personalized feedback to help improve levels in sports such as golf and tennis. By comparing with a vast amount of athlete data, the system achieves automation and data-driven insights.

Reddit, known for its vast user community and rich data resources, has opened up new revenue streams through data licensing. The platform hosts over 100,000 active communities, attracting 76 million users daily and generating a massive amount of data that includes real-time discussions, opinions, and interactions, which is highly valuable for AI and machine learning model development. With the rapid advancement of AI technology, Reddit's data has become a key resource for training AI models. In 2023, Reddit launched its data licensing program, which is expected to grow significantly in 2024.

Noetik, a biotechnology company focused on artificial intelligence, announced that it has successfully completed a $40 million Series A funding round. The round was led by Polaris Partners, with participation from new investors including Khosla Ventures, Wittington Ventures, Breakout Ventures, as well as existing investors DCVC, Zetta Venture Partners, and Catalio Capital Management.

Turkish high school student Selin Alara Ornek innovatively created an AI-driven machine guide dog using Nvidia technology to provide high-tech assistance for visually impaired people. Combining advanced robotics and AI technology, Ornek's IC4U project utilizes AI cameras and sound sensors for precise object detection and recognition, aiding in daily activities such as shopping. This quadruped robot, powered by the Nvidia Jetson Nano chip, efficiently processes visual and auditory information. Ornek is committed to...

During the themed event on 'The Deep Empowerment of AI Technology in Primary and Secondary Education', Li Ying, CEO of Baidu Xiao Du Technology, shared the latest developments in Xiao Du's education business. As of July 2024, the Xiao Du Learning Machine has surpassed 2 million cumulative users, with more than 50% being daily active users, indicating its popularity. The usage rate of AI teachers has increased by 3.5 times compared to traditional methods, with users spending an additional 25% time on the learning machine, totaling 114 minutes per day. Notably, during the immersive interactive preview section for Chinese language, the average usage time reached 22 minutes.

Geekbench AI is a brand new cross-platform tool designed to evaluate device performance under AI-intensive workloads. It tests the CPU, GPU, and NPU to determine the device's ability to handle machine learning applications. It supports various AI frameworks such as ONNX, CoreML, TensorFlow Lite, and OpenVINO, and evaluates performance from both speed and accuracy dimensions, providing scores in full precision, half precision, and quantized precision. Additionally, Geekbench AI's testing results...

Geekbench AI is a new cross-platform tool designed to evaluate the performance of devices under compute-intensive workloads, with a particular focus on machine learning tasks. It provides full precision, half precision, and quantized scores by measuring the performance of CPUs, GPUs, and NPUs, and includes accuracy assessments. It supports various machine learning frameworks, including ONNX, CoreML, TensorFlow Lite, and OpenVINO. This tool is available on Windows, macOS, Linux, Android, and iOS.

The report indicates that global shipments of AI-capable PCs reached 8.8 million units in the second quarter of 2024, with Apple's Mac accounting for 60%, maintaining a leading position in the AI-enabled PC market. In the over $800 market segment, AI-capable PC shipments for Windows increased from 7% in the first quarter to 14%. AI capability is seen as the new standard for PCs, with dedicated AI processing chips provided by AMD, Apple, Intel, and Qualcomm. AI-enabled PCs refer to desktop or laptop computers equipped with AI processing chips, such as Apple.

Sakana AI has announced the launch of the 'AI Scientist' system, a comprehensive AI platform capable of independently conducting scientific research, from concept generation, coding, experiment execution, result summarization to paper writing, including peer review. Its design aims to mimic the human scientific research process. The system has demonstrated innovative contributions in several subfields of machine learning, such as diffusion models, transformers, and deep understanding, showcasing high computational efficiency and low costs (approximately $15 per paper). Although the initial version of the system may have some limitations.

Developer Evan Zhou successfully manipulated Apple's new AI system, Apple Intelligence, in MacOS 15.1 Beta 1 using an attack method called 'prompt injection'. Initially aimed at rewriting text quality, Zhou constructed an attack that overwrites system prompts by leveraging prompt information shared by Reddit users, causing the AI to ignore original instructions and respond with new text. This discovery has garnered widespread attention, although Apple Intelligence...

The T-MAC (Table-Lookup-based MAC) technology aims to address the memory and computational constraints of deploying large language models (LLMs) on edge devices. By quantizing model weights into low-bit representations, T-MAC replaces traditional multiplication operations with look-up tables (LUT), significantly enhancing running efficiency on CPUs. This method greatly reduces the memory required for calculations, enabling billion-parameter LLMs to run efficiently on resource-constrained devices for smart upgrades. Compared to existing implementations, T-

Hugging Face has acquired XetHub to accelerate the process of AI democratization. Founded by former Apple engineers, XetHub focuses on simplifying the building of machine learning and AI applications. The team secured $7.5 million in funding from Madrona Ventures in 2021 to accelerate their development. This acquisition signifies Hugging Face, an influential platform in the AI development space, further expanding its technological capabilities and talent pool. The inclusion of XetHub reflects the industry's growth.

OpenAI has released a ChatGPT desktop application for macOS, allowing users to conveniently use the AI chatbot while multitasking. Key highlights include improved multitasking features that enable users to invoke the ChatGPT search box via shortcuts without interrupting other applications, expandable views for generated tables and charts to support user interaction and downloads. Additionally, the app introduces several new keyboard shortcuts, customization options, and performance security enhancements. In case of issues with missing new features after download,

OpenAI has updated its ChatGPT application for Mac, introducing a split window feature that allows users to interact with ChatGPT while keeping the application interface and mini ChatGPT window displayed side by side by pressing Option + Space. This enhances multitasking efficiency by reducing the hassle of switching windows, especially convenient when using small screens. Additionally, OpenAI is exploring a text generation feature aimed at...

MedSAM-2, based on the Segment Anything Model 2 (SAM2) framework, brings revolutionary changes to the field of medical image segmentation. Its core breakthrough treats 2D and 3D images as video sequences, enabling efficient segmentation of three-dimensional images. The 'one-click segmentation' feature allows users to specify a target in a single image, and the model can automatically recognize and accurately segment the same type of object in subsequent images without the need for continuous input. As a general model, it can handle any image.