Discover the Latest AI News📰

Alibaba has unveiled Qwen3.6-35B-A3B, an open-source AI model that punches above its weight in programming tasks. Despite only activating 3 billion parameters at a time, this 'mixture of experts' model outperforms larger rivals while using less computing power. It shines in coding assistance, spatial reasoning, and visual understanding - already matching some premium AI services. Developers can now tap into this efficient brainpower through Alibaba's cloud platform.

Tongyi Yunqi, the company behind Alibaba's Tongyi Qianwen AI model, has significantly expanded its financial base with a 100% capital increase to 200 million yuan. The Hangzhou-based firm now counts both Alibaba Cloud and newly-added shareholder Hangzhou Tongyi Lab among its owners. Established in 2018, the company has steadily grown its operations in AI-related software and hardware sectors.

A domestic dispute in China's Qinghai province took a bizarre turn when a man used AI to create fake suicide photos to scare his wife. The images, showing him in the Yellow River, triggered a full-scale police search before authorities discovered the hoax. Now facing administrative detention, the case highlights growing concerns about misuse of AI technology in personal conflicts.

ByteDance is rolling out an ambitious recruitment drive targeting doctoral students worldwide for cutting-edge AI roles. The program offers both full-time positions and internships across multiple tech hubs, prioritizing candidates with strong research backgrounds or competition wins. From large language models to AI safety, the company is casting a wide net for tomorrow's tech leaders.

In a major power move, OpenAI is investing over $20 billion in Cerebras Systems for AI chip technology, with the potential to acquire up to 10% equity. The deal expands their existing partnership and could total $30 billion when including data center development funding. This strategic play comes as Cerebras prepares for a possible IPO later this year, highlighting the intensifying race for AI computing power.

Silicon Valley's battle for AI talent reaches new heights as Meta poaches yet another star engineer from Thinking Machines. Joshua Gross, architect of the startup's flagship Tinker software, recently jumped ship to lead Meta's Super Intelligence Lab team. This marks at least the sixth high-profile departure from the $12 billion startup, highlighting the fierce competition among tech giants for limited AI expertise. While companies offer nine-figure compensation packages to lure top minds, startups like Thinking Machines fight back by recruiting industry heavyweights of their own.

NVIDIA's research team has unveiled Lyra 2.0, an advanced 3D scene generation system that builds expansive virtual environments from just one photo. The technology can create coherent 90-meter digital landscapes while solving traditional distortion issues. Benchmark tests show Lyra 2.0 outperforms competitors in image quality and camera control, with its fast version offering 13x better efficiency. The system integrates seamlessly with physical engines like Nvidia Isaac Sim, opening new possibilities for robotics training and AI development.

Anthropic's new mandatory ID verification for Claude AI has users up in arms. The process requires holding physical IDs for photos, with data handled by third-party Persona. But many see it as a prelude to account suspensions - like one teen developer who lost access despite paying for services. With privacy concerns over data sharing and strict age limits, Claude's security measures are creating as many problems as they solve.

Google has introduced a game-changing update to its Gemini AI, allowing it to tap directly into users' personal photo libraries. When enabled, the Personal Intelligence feature can access your Google Photos to create custom AI-generated images – imagine transforming family vacation snaps into animated masterpieces with a simple request. This breakthrough blurs the line between private memories and AI creativity, offering truly personalized digital art.

Starbucks is testing a ChatGPT-powered feature that acts like your personal coffee concierge. Just tell the app how you're feeling - tired, stressed, or needing energy - and get customized drink recommendations. While promising more personalized orders, some experts caution about becoming too reliant on algorithmic suggestions for simple daily choices.

OpenAI has unveiled GPT-Rosalind, a specialized AI model for life sciences research. Named after DNA pioneer Rosalind Franklin, this tool helps researchers analyze biochemical data, generate hypotheses, and accelerate drug development. Currently in limited preview with partners like Amgen and Moderna, it outperforms humans in some predictive tasks while raising questions about AI's role in scientific research.

In a significant shift for the AI landscape, women have surpassed men as the majority users of ChatGPT, marking a cultural turning point in technology adoption. The latest data reveals female users now represent over 50% of the platform's near-billion user base, a dramatic change from the 80-20 male dominance at launch. Meanwhile, OpenAI's computing power expansion signals aggressive plans for future AI development.

Chinese tech company iFLYTEK has unveiled a significant upgrade to its AstronClaw AI system, introducing nine new products and a novel 'software-hardware integrated' AI Agent architecture. The upgrade transforms AI assistants from mere conversational partners into active participants in physical tasks, with applications ranging from office productivity to wearable tech. A standout product is the lightweight GlassClaw AI glasses, which combine voice and visual recognition to interact with environments intelligently.

In a major power play for AI hardware dominance, OpenAI has committed over $2 billion to Cerebras Systems in a three-year chip supply deal that doubles their previous agreement. The partnership goes beyond chips - OpenAI is investing $1 billion to help build Cerebras' data center infrastructure while securing warrants for up to 10% ownership. Meanwhile, Cerebras eyes a blockbuster IPO valuing the company at $35+ billion, looking to raise $3 billion as demand for specialized AI processors skyrockets.

OpenAI's ChatGPT has reached a major milestone with nearly 1 billion weekly active users. In a striking shift, women now make up over 50% of users for the first time - up from just 20% when the platform launched. This reflects AI's growing mainstream appeal across demographics. Meanwhile, OpenAI's computing power is skyrocketing, expected to grow nearly tenfold by 2025 to support this expanding user base.

The retail giant Miniso is betting big on artificial intelligence, creating a dedicated division to develop smart tools for everything from store locations to supply chains. This move puts Miniso at the forefront of retail tech, using AI to make sharper business decisions and streamline operations globally. They're already hunting for top tech talent to bring their digital vision to life.

Lenovo has unveiled Tianxi AI Pro, an advanced AI assistant tailored for government and enterprise use. What sets it apart? The system shifts from tool-centric to task-oriented operations, offers a unique dual interface, and prioritizes security with local data processing. It can handle everything from meeting coordination to document generation - all while keeping sensitive information safely offline.

OpenAI has taken a bold step into life sciences with GPT-Rosalind, a specialized AI model named after DNA pioneer Rosalind Franklin. Unlike general chatbots, this tool can analyze protein structures, predict gene functions, and suggest experimental pathways—outperforming human experts in some tests. Currently available only to select biotech firms, it promises to accelerate drug development while raising important questions about AI's role in scientific discovery.

OpenAI just gave its Codex AI assistant some serious upgrades. The tool can now control Mac applications independently, run multiple tasks simultaneously, and remember your workflow preferences for days. Imagine having a digital assistant that clicks, types, and browses for you - that's what Codex delivers with this update. Developers will particularly love how it seamlessly picks up paused projects and even suggests next steps.

Anthropic's Chief Product Officer Mike Krieger has stepped down from Figma's board, sparking industry speculation about the AI firm's growing ambitions in design tools. The move comes as Anthropic prepares to launch its Opus4.7 model with native design capabilities, potentially positioning it as a direct competitor to Figma. This development has sent ripples through the tech sector, raising questions about AI's growing encroachment on traditional software domains and the future of specialized design platforms.