What Is Frozen v2, Google’s Reported Gemini-Only AI Chip?

By Devanshi Kashyap | July 21, 2026

Google is reportedly developing a new AI chip designed specifically for its Gemini AI models. Unlike a new chatbot or software update that users can interact with directly, this development is focused on the hardware powering AI. The reported chip, internally named Frozen v2, is aimed at making Gemini faster, more efficient and less expensive to run. The move also reflects a broader shift in the AI industry. Over the past few years, the focus has mostly been on building more capable AI models. Now, companies are investing in the chips that run those models, as the cost and computing […]

The US-China AI Fight Is No Longer Just About Hardware

By Khwaish Manwani | July 21, 2026

China is extensively focused on succumbing to artificial intelligence as a strategic asset. The newest discussions around strict controls suggest Beijing is no longer restricting chips or AI models. It is now focusing on the people, data, model weights, and intellectual property that make those systems feasible. The shift puts forward a new possibility where China views AI leaks and talent poaching as safety risks rather than corporate disputes. Chinese regulators are now planning to impose new export controls covering frontier AI semiconductor technologies. The conversations reportedly involve the Ministry of Commerce alongside native domestic AI firms including Alibaba, ByteDance, […]

Grok Is Now Inside Excel, on Microsoft’s Own Turf

By Abhijay Singh Rawat | July 20, 2026

SpaceXAI has brought Grok into Microsoft Excel. Announced today, the free Microsoft 365 add-in lets users ask questions about their data in plain English, write formulas from a description, build charts, and run forecasts without leaving the workbook. It arrived alongside versions for Word and PowerPoint, so this is not a single Excel tool but Grok moving into Microsoft’s entire Office suite. SpaceXAI says users can work without exporting tables or copying data into a separate chatbot, since Grok runs in the sheet already open. It is a genuinely useful capability. It also raises an obvious question, because the app […]

SK Group Says Nations May Soon Compete for AI Memory

SK Chief Warns the AI Memory Shortage Could Turn Geopolitical

By Khwaish Manwani | July 20, 2026

For the past few years, artificial intelligence has been defined by GPU access. Companies and governments compete to secure Nvidia accelerators because they power frontier AI models. That is about to change. According to SK Group Chairman Chey Tae-won, the next bottleneck is not the processors, but the memory chips that let those processors work. Speaking this week, Chey warned that demand for AI memory is growing so fast it could pull governments in, turning what was once a supply-chain issue into a question of economic security. As nations race to build their own AI capabilities, securing memory is becoming […]

Kimi K3 Was So Popular Moonshot Had to Stop Selling It

By Khwaish Manwani | July 20, 2026

Moonshot AI’s new venture has created a problem for consumers. Just days after revealing Kimi K3, the Chinese startup halted new paid subscriptions after demand surpassed supply. Existing subscribers keep uninterrupted access, but anyone hoping to buy a membership is now turned away until more capacity becomes available. The move highlights an emerging reality in AI. Companies are no longer competing only to build capable models. They also have to scale up the computing power to keep those models running. When that infrastructure falls short, it is users, not companies, who lose out. What Does the Kimi K3 Pause Mean […]

Qwen 3.8 Claims Second Only to Fable 5, but Shows No Benchmarks

By Khwaish Manwani | July 20, 2026

Alibaba has revealed Qwen 3.8, a novel multimodal artificial intelligence model with 2.4 trillion parameters. This makes it among the world’s largest open-source AI systems. The company says the model is second only to Anthropic’s Fable 5, although it has yet to reveal independent benchmarks supporting the claim. But beyond the eye-catching parameter count, the question put forward is whether size alone reflects efficiency, because Qwen 3.8 uses Mixture of Experts (MoE) infrastructure. Only a fragment of those 2.4 trillion parameters are actually used for one request. That shifts the conversation from how large the model is to how capable […]

A Model Called Honeycomb Leaked in Cursor. Fans Say It’s Opus 5.

By Devanshi Kashyap | July 17, 2026

A new model name called Honeycomb has become a major topic of discussion in the AI community after it reportedly appeared inside Cursor’s configuration files. Many developers and AI enthusiasts believe this could be Anthropic’s next flagship AI model, Claude Opus 5. However, there is an important point to remember. The Honeycomb leak is not an official announcement from Anthropic. It is simply a model name that appeared in Cursor, along with a few leaked details. Everything beyond that, including the model’s final name, features, and launch date, is based on speculation, so readers should separate the confirmed information from […]

Indonesia Moves to Become Southeast Asia’s First Nation to Write AI Into Copyright Law

By Khwaish Manwani | July 17, 2026

Indonesia is unveiling one of the biggest reforms for the artificial intelligence niche. This could make it the first Southeast Asian country to define how artificial intelligence incorporates into the copyright law. Instead of deeming artificial intelligence as an author, the draft legislation aims to draw a legal boundary. AI can help in creating the content, but the rights should remain with humans. The proposal comes amidst federal authorities worldwide struggle to balance innovation with creator rights. This makes Indonesia an early test case for whether copyright law can govern AI without granting authorship to machines. Why Is Indonesia’s AI […]

Google Delays Gemini 3.5 Pro Launch as Coding Capabilities Fall Short of Internal Standards

By Khwaish Manwani | July 17, 2026

Google’s plan to halt the public release of Gemini 3.5 Pro emphasizes how arduous the artificial intelligence race is. The model was first revealed at Google I/O in May, with better reasoning, multimodal abilities, and industrial leading context length, and was expected to launch in June. Instead, a new report suggests that Google is stopping its flagship AI because its coding capabilities did not meet internal checks. The delay comes at a time when competitors including OpenAI, Meta, Anthropic, and xAI are readying new coding-focused AI models, making development one of the most important breakthroughs in the AI niche. Why […]

Is Kimi K3 World’s Largest Open AI Model? Here’s What We Know

By Devanshi Kashyap | July 17, 2026

Moonshot AI has launched Kimi K3, which they claim is the world’s largest open AI model. The announcement is another major development in the global AI race as the Chinese startup is aiming to narrow the gap with leading U.S. AI companies. Unlike closed AI models, Kimi K3 is released as an open weight model, which gives developers more freedom to use, customize and deploy it for different applications. While leaks suggested 2.5 trillion parameters, the mode is now confirmed to be built using a 2.8 trillion parameter Mixture-of-Experts (MoE) architecture and supports a 1 million token context window. These […]

AI & Tech News