JD Logistics has debuted its self-developed unmanned light truck, the JD Logistics VAN, featuring L4-level autonomous driving capability. With a 24-cubic-meter cargo capacity and 400km range, it aims to revolutionize logistics operations by cutting costs by 60% and improving efficiency in e-commerce, manufacturing, and cold chain sectors.
Silicon Valley startup Cluely has doubled its annual recurring revenue to $7 million within a week of launching a new enterprise AI product. The tool provides real-time meeting assistance but faces competition from open-source alternatives.
Tencent's intelligent assistant Yuanbao has upgraded its search functionality, now displaying images and videos alongside text results. The 'Search More with One Sentence' feature allows users to receive multimedia answers to queries, improving learning and problem-solving experiences. The update integrates video tutorials and visual guides for tasks ranging from skill acquisition to everyday life hacks.
Meta is developing proactive chatbots that message users first, rather than waiting for prompts. These customizable bots, built via AI Studio, remember preferences and follow up within 14 days. While aimed at deepening engagement, concerns linger over privacy and misuse risks.
Hitachi Energy CEO warns that AI data centers' erratic power demands could destabilize global electricity grids. With consumption projected to double by 2030, governments must implement regulations to manage these fluctuations while the industry addresses transformer shortages through major investments.
Google has launched its Veo 3 video generation model globally, now available in 159 countries for Gemini users. The AI tool creates 8-second videos from text prompts but is currently limited to paid AI Pro subscribers with a daily cap of three generations. Future updates may include image-to-video conversion capabilities.
Figma has filed for a NYSE IPO with a $20B valuation, marking a major milestone for the cloud-based design platform. The company shows strong financial growth and AI integration, positioning itself as a leader in collaborative design tools while expanding into enterprise and global markets.
WeChat Pay has introduced its Merchant Connect Platform (MCP), integrating AI with payment processing to create new revenue streams. The feature enables real-time payments during AI interactions, improving conversion rates and providing merchants with actionable data. This innovation marks a significant step in AI commercialization.
ByteDance has open-sourced Trae-Agent, the core component of its AI-native IDE Trae, offering developers worldwide an intelligent programming assistant. The tool supports natural language-driven automation, multi-model compatibility, and advanced debugging features, aiming to revolutionize development workflows. With 10M+ monthly users, this move strengthens ByteDance's position in AI-driven developer tools.
Meitu WHEE has launched a groundbreaking 'One-Sentence Photo Editing' feature, enabling users to achieve complex photo edits through simple voice or text commands. The AI-driven tool simplifies style changes, local modifications, and text processing, eliminating the need for manual adjustments. The feature is available on the web version at www.whee.com.
Meta is internally testing a new feature called 'Omni' that enables AI chatbots to initiate conversations without user prompts. The feature aims to enhance engagement by recalling past interactions, with strict limits on message frequency. Meta anticipates significant revenue growth from its generative AI products by 2025.
TNG's enhanced DeepSeek-TNG-R1T2-Chimera model achieves a 200% improvement in inference efficiency and reduces costs using the innovative AoE architecture. This hybrid model outperforms standard versions in benchmark tests, showcasing advanced reasoning capabilities and cost-effectiveness.
Xpeng Motors has launched the G7 Ultra, featuring groundbreaking 'VLA+VLM' AI models for autonomous driving and smart cabin functionality. The vehicle boasts Qualcomm 8295P and Turing AI chips, offering 26x more computing power than industry standards. Key features include adaptive routing, hazard detection, and pet monitoring, with full functionality rolling out later this year.
JPMorgan predicts Baidu's market cap could surpass $79.6 billion amid growing Robotaxi industry momentum. With Waymo's valuation projected at $75 billion and Baidu's Apollo Go expanding globally, autonomous driving is reaching a critical inflection point, drawing comparisons to the 'ChatGPT moment' by Bank of America.
A federal judge has granted The New York Times access to OpenAI's user logs, including deleted content, as part of an ongoing copyright infringement lawsuit. The newspaper argues this data is crucial to prove ChatGPT trained on copyrighted material. OpenAI opposes the ruling, citing privacy concerns.
Ambiq Micro, a Texas-based chip design firm, has filed for a U.S. IPO, reporting a 16.1% sales growth in 2024 fueled by generative AI demand. Despite narrowing losses, the company faces customer concentration risks while positioning itself in the energy-efficient 'edge AI' market.
A former OpenAI researcher, Lucas Beyer, has publicly stated that neither he nor his colleagues received the $100 million signing bonus Meta reportedly offered. The revelation raises questions about Meta's recruitment strategies amid its aggressive AI talent acquisition push. OpenAI CEO Sam Altman has criticized Meta's approach, doubting its ability to foster positive corporate culture.
A KPMG China report reveals China dominates 70% of global medical large model releases, with AI-driven healthcare innovations accelerating. The sector is projected to grow significantly by 2025, fueled by policy support and tech advancements.
Kyutai Labs has open-sourced its Kyutai TTS technology, offering low-latency, high-fidelity voice synthesis. The system supports streaming text input and generates precise word timestamps, making it ideal for real-time applications. Currently supporting English and French, it achieves high accuracy with word error rates below 3.3%.
Google has released Veo 3, its advanced AI video generation model, exclusively for Pro and Ultra subscribers. The tool offers HD quality, synchronized audio, and multi-modal creation. A photo-to-video feature is coming soon.
