Discover the Latest AI News📰

BuzzFeed's much-hyped AI pivot hit a snag at SXSW when its new apps Conjure and BF Island received awkward silence instead of applause. The digital media company, struggling financially after cutting its Pulitzer-winning newsroom, faces growing skepticism about its AI-first strategy. While CEO Jonah Peretti insists content is no longer king, the lukewarm reception suggests users still crave substance over gimmicks.

Forget expensive app developers - young entrepreneurs are now creating custom AI applications for pocket change. Platforms like Xianyu and Xiaohongshu have become marketplaces for 'digital sole proprietors' offering everything from bug fixes to full app creation, all priced between 5-20 yuan. These tech-savvy individuals are turning everyday problems into profitable micro-businesses, proving you don't need coding skills to succeed in today's AI-driven economy.

Alibaba Cloud is raising prices for its AI computing and storage services by up to 34%, reflecting growing global demand and supply chain pressures. The increases affect key products like the Zhenwu 810E computing cards and CPFS storage solutions, as the company prioritizes resources for booming large model applications. This move signals how the AI industry's explosive growth is reshaping cloud infrastructure economics.

ByteDance has introduced ByteClaw, a new security tool designed to manage employee access to large AI models within its internal networks. The company also released comprehensive guidelines addressing five major security risks in AI applications. These measures aim to balance innovation with safety as AI tools become more prevalent in corporate environments. The move sets a benchmark for how tech firms can securely integrate powerful AI systems into their operations.

MiniMax has unveiled M2.7, China's first large language model capable of actively participating in its own development. This breakthrough allows the AI to build testing frameworks and optimize its performance without human intervention. Meanwhile, the AI industry sees significant movements, with LuChen Technology securing major funding and AliCloud adjusting prices due to surging demand. As models begin 'raising themselves,' we're witnessing a new chapter in artificial intelligence evolution.

At NVIDIA's GTC2026 conference, Moonshot AI founder Yang Zhilin unveiled the three pillars powering their Kimi K2.5 model - token efficiency, long context memory, and agent clusters. This marks a shift from brute-force computing to smarter architecture design. The approach could redefine how we measure AI intelligence beyond just parameter counts.

Tencent has upgraded its AI assistant QClaw with deeper WeChat integration, now available as a Mini Program. Users can seamlessly transfer files between devices and soon enjoy voice and image interactions. The update includes an 'Inspiration Square' with pre-loaded tasks for work and play. While AI assistants are becoming ubiquitous, experts warn about security risks that come with their advanced capabilities.

Unsloth AI has launched a game-changing platform that lets developers fine-tune large language models right on their personal computers. The new Unsloth Studio cuts GPU memory needs by 70% while doubling training speeds, making powerful AI development accessible without expensive cloud services. With its intuitive visual interface, the tool eliminates complex setup processes and brings professional-grade model customization within reach of individual creators and small teams.

Baidu's search engine is getting a major AI upgrade. The company has transferred a top algorithm expert to lead its mobile app development, signaling deeper integration of large language models into search and recommendations. This shift could fundamentally change how we interact with Baidu's services, moving beyond simple keyword matching to understanding user intent.

As AI actors gain traction in minor film roles, industry veteran Yu Zheng pushes back, arguing technology still can't replicate the emotional depth of live performances. While AI offers cost savings for action scenes and short-form content, its limitations in creating genuine human connection remain stark. The debate highlights growing tensions between technological efficiency and artistic authenticity in entertainment.

Six major tech companies—Anthropic, Amazon, GitHub, Google, Microsoft, and OpenAI—have pooled $12.5 million to help open-source developers combat the flood of low-quality AI-generated security reports. The funding will support tools and processes to filter out false alarms, allowing maintainers to focus on genuine threats. This initiative responds to growing concerns about AI automation overwhelming open-source projects with what developers call 'vulnerability noise.'

Tencent's latest financial results reveal strong momentum in its enterprise services division, with B2B revenue hitting 229.4 billion yuan - a new high for the tech giant. The company credits its AI investments and cloud computing improvements for driving this growth, while also announcing upcoming upgrades to its HY AI models. With products like WorkBuddy gaining traction, Tencent appears well-positioned in the competitive AI services market.

Tencent's gaming division hit new heights in 2025, with revenue jumping 22% to 241.6 billion yuan. International markets proved particularly lucrative, surpassing $10 billion for the first time thanks to hits like PUBG Mobile. The company's AI investments are transforming gameplay, with over 110 million players engaging with intelligent NPCs. Domestic favorites like Honor of Kings continue to thrive while new titles like Delta Force show promising growth.

Microsoft is threatening legal action against OpenAI over its $50 billion cloud partnership with Amazon, alleging it violates their exclusive agreement. The dispute centers on OpenAI's new enterprise product Frontier and technical workarounds designed to bypass Microsoft's terms. This conflict comes at a delicate time for OpenAI, already facing legal challenges from Elon Musk while preparing for a potential 2026 IPO.

In a rare show of unity, Google, Microsoft, OpenAI and other tech heavyweights have pooled $12.5 million to help the Linux Foundation tackle a growing problem - the flood of unreliable AI-generated security reports overwhelming open-source projects. The funding will support efforts to filter out low-quality vulnerability alerts while protecting against real threats. This collaboration marks another step in the industry's push to safeguard the digital infrastructure we all depend on.

Japan's Rakuten Group finds itself in hot water after releasing its AI 3.0 model without proper attribution to the open-source DeepSeek-V3 it was built upon. The tech community cried foul when the required MIT license disappeared from the package, forcing Rakuten to scramble with corrective measures. While legally compliant now, questions remain about transparency and proper use of public funds for what critics call 'repackaged' technology.

Wang Xingxing, founder of Autonomous Intelligent Technology, predicts embodied AI won't reach its breakthrough moment for another 2-3 years. While progress continues, robots still struggle with complex real-world tasks. The industry awaits that pivotal threshold when machines can handle unfamiliar scenarios as smoothly as ChatGPT processes text.

Beijing has launched a targeted campaign to clean up AI misuse online. The one-month initiative aims to tackle everything from deepfake scams to AI-generated pornography, focusing on five key problem areas. Authorities will work with platforms to strengthen content moderation while cracking down on illegal services that exploit AI technology.

Didi has officially launched its AI travel assistant 'Xiao Di' after six months of beta testing. The smart assistant understands natural language requests like 'I feel carsick' or 'pick up my friend first', automatically matching them with appropriate services. With over 90 service tags, it aims to simplify complex travel needs into one-step solutions. Users can now upgrade their Didi app to try this conversational approach to booking rides.

Tencent's AI workbench IMA just leveled up with its new 'Skills' feature, making automation smoother than ever. The update lets popular tools like OpenClaw and WorkBuddy play nice with IMA's inner workings through simple API connections. Right out the gate, users get a handy Notes skill for all their reading, writing, and searching needs - with a Knowledge Base skill waiting in the wings. This move shows how AI tools are evolving from solo acts to team players in our digital workflows.