SenseTime recently launched its new app, 'Magic Photo Capture', providing users with a creative and fun AI imaging experience. With the arrival of the Lunar New Year, this app aims to help users create personalized New Year photos using advanced artificial intelligence technology, enhancing the festive atmosphere. 'Magic Photo Capture' is not just an ordinary photography app; it includes a variety of templates in its mini-program, allowing users to choose from graduation photos, creative group photos, classic art style portraits, travel photos, new national style, and more.

At the latest JP Morgan Healthcare Conference in 2025, NVIDIA and Innophore announced an exciting collaboration to launch an AI-driven drug safety screening platform named CavitOmiX by the end of January 2025. This platform aims to improve the efficiency and accuracy of drug design through the analysis of protein binding sites, marking a revolutionary transformation in the field of drug development. Christian Gruber, CEO of Innophore.

Globally, various enterprises are striving to understand generative AI and explore the potential value it may bring. According to Deloitte's recent report on the state of generative AI, most organizations are performing well in achieving returns on their investments. This report is based on a survey of 2,773 leaders and covers the business conditions of companies in 14 countries, revealing the progress and challenges faced by organizations in the application of generative AI. Image source note: The image is generated by AI and authorized by Midjourney.

A significant shift has occurred in the partnership between Microsoft and OpenAI. With OpenAI announcing the signing of a Stargate agreement with companies like SoftBank and Oracle, Microsoft is no longer the exclusive data center infrastructure provider for OpenAI. According to the new agreement, Microsoft has been granted a 'right of first refusal' on new cloud computing capacity from OpenAI, meaning that if Microsoft cannot meet demand, OpenAI can seek support from other cloud service providers. Microsoft confirmed this change in a blog post, stating that OpenAI has turned to Azure.

Perplexity announced on Tuesday the launch of its enterprise-level API service "Sonar", allowing businesses and developers to integrate its generative AI search capabilities into their own applications. The service is available in two versions: the basic Sonar and the advanced Sonar Pro, and it supports customers in customizing the information sources of the AI search engine. Unlike traditional AI models that rely solely on training data, Perplexity emphasizes that Sonar can connect to the internet in real-time and obtain information from reliable sources, thereby providing more factual and authoritative results.

During today's Baidu AI Open Day event, Baidu's Vice President and Head of the Wenku and Baidu Cloud Business Units, Wang Ying, revealed the latest developments in the application of AI technology by Baidu Wenku. It was introduced that the monthly active users of Baidu Wenku's AI features have surpassed 90 million, with daily active users increasing by 230% year-on-year, and the number of paid users exceeding 40 million, demonstrating the platform's strong appeal in the application of AI features. In the past year, Baidu Wenku has added over 100 AI features, including various innovative tools to meet user needs in document processing.

The Dark Side of the Moon Company proudly launches its brand new SOTA (state-of-the-art) model—k1.5 Multimodal Thinking Model, marking a significant breakthrough in multimodal reasoning and general reasoning fields. This model not only possesses outstanding multimodal processing capabilities but also demonstrates exceptional general reasoning abilities, effectively addressing various complex tasks. The standout feature of the k1.5 model is its multimodal reasoning capability. It can process information from different modalities including text, images, and sounds simultaneously, providing more comprehensive and accurate results.

Startup Friend plans to launch a $99 AI smart necklace intended to be a digital companion for users; however, the company has announced a delay in shipping the first batch of products, now expected in the third quarter. The original plan to ship to pre-order customers in the first quarter now seems unfeasible. Co-founder and CEO Avi Schiffman stated in an email to customers: 'While I hoped to ship in the first quarter of this year, I still need to refine the product further. In fact,

Google Research recently announced the innovative 'Titans' series model architecture, achieving a groundbreaking 2 million token context length through bionic design, with plans to open-source related technologies in the future. The core innovation of this architecture is the introduction of a Deep Neural Long-Term Memory Module, inspired by the human memory system. Titans cleverly combines the rapid response capability of short-term memory with the persistence characteristics of long-term memory, while utilizing an attention mechanism to handle immediate context, forming an efficient information processing system.

As global competition in the field of artificial intelligence intensifies, domestic AI models are rapidly emerging. Recently, the three domestic AI contenders—DeepSeek, Kimi, and MiniMax—successfully launched new versions, demonstrating potential to compete with OpenAI's GPT series. On January 20, DeepSeek officially released DeepSeek-R1 and simultaneously open-sourced the model weights. DeepSeek-R1 heavily utilized reinforcement learning techniques during its post-training phase.

On his first day in office, President Trump announced the revocation of an executive order signed by former President Biden in 2023, which aimed to mitigate the risks posed by artificial intelligence (AI) to consumers, workers, and national security. Biden's executive order directed the National Institute of Standards and Technology (NIST), under the U.S. Department of Commerce, to establish guidelines to help businesses identify and rectify flaws in their AI models, including biases. Furthermore, the executive order required developers of AI systems to report to the U.S. government before publicly launching their products.

Recently, Mita AI Search officially launched its brand new 'Reading Mode' feature, addressing the inconvenience users face when reading PDF documents on mobile devices. This feature has been meticulously developed over the past six months and is now available to users. Since the launch of 'Library Search', Mita AI Search has helped users conveniently find a wealth of valuable PDF documents. However, these documents often suffer from messy formatting and readability issues when viewed on mobile. To improve this situation, the Mita AI Search team began development six months ago.

Recently, Zhejiang University and Alibaba DAMO Academy jointly released a remarkable study aimed at creating high-quality multimodal textbooks from educational videos. This innovative research not only provides new ideas for training large-scale language models (VLMs) but may also change the way educational resources are utilized. With the rapid development of artificial intelligence technology, the pre-training corpus of VLMs mainly relies on visual-text pairs and visually intertwined data. However, much of this current data comes from the web, where the correlation between text and images is weak, and the knowledge density is relatively low.