Sanjay Sahay

Money should go where the mouth is and SoftBank has clearly understood this investment strategy. It has started making moves in tune with the growth of the AI industry. SoftBank Group Corp. has transformed its portfolio, with AI-related holdings surging from 18% of net asset value three years ago to over 60% today, fueled by […]

Google has unleashed an upgraded Deep Think reasoning mode in Gemini 3, smashing benchmarks and stealing the spotlight from OpenAI and Anthropic in 2026. It crushes ARC-AGI-2 at 84.6%, far ahead of Opus 4.6’s 68.8% and GPT-5.2’s 52.9%, while hitting 48.4% on Humanity’s Last Exam — a new record. Deep Think earns gold-medal scores on

Elon Musk’s xAI, fresh off its SpaceX merger, hosted its first all-hands meeting amid a wave of key founding team departures, including co-founders like Jimmy Ba and Tony Wu. Musk openly acknowledged the exits but framed the subsequent reorganization as essential for greater effectiveness at scale,* signaling no slowdown in ambitions. The new structure streamlines

xAI, Elon Musk’s hot AI startup, is losing its top brains fast. Just days after merging with SpaceX, two big co-founders —Tony Wu and Jimmy Ba— quit. Wu led Grok’s smart thinking tech; Ba was a key player too. Now five founders have gone in under a year. What’s cooking behind the scenes? Musk built

China’s AI labs have surged ahead in video generation, building on rapid releases like Kuaishou’s Kling series since 2024 and ByteDance’s own Seedance 1.0 in 2025. Seedance 2.0, now in beta on Jimeng AI, handles text, images, videos, and audio inputs for 2K cinematic clips with synced sound, outpacing rivals in multimodal control. This launch

Jeff Bezos and other billionaires are betting big— $3 billion big—on Altos Labs, a secretive biotech startup aiming to reverse aging at the cellular level. Launched in 2022, it’s one of the richest ventures in longevity science, pulling in top experts from academia and pharma. The core idea? Epigenetic reprogramming. Scientists use special proteins called

Anthropic has released Claude Opus 4.6, its most advanced model yet, excelling in agentic coding, long-context reasoning, and professional tasks like financial analysis. This upgrade brings significant improvements in planning, debugging, and handling large codebases compared to Opus 4.5. A standout feature is “agent teams” in Claude Code, allowing multiple AI agents to collaborate in

In what shape and geography the big tech battle would be fought no one knows. Besides the companies the governments are getting badly embroiled into it. French authorities raided X’s Paris offices this week, probing failures to remove child sexual abuse images and content denying crimes against humanity. This escalates EU fines of €120M and

OpenAI just launched GPT-5.3-Codex, a super-smart coding tool that’s like a brainy robot programmer. It combines top skills in writing code and solving problems into one fast package. Even better, early versions of this AI helped fix bugs in its own training, manage its rollout, and check results—making it smarter on its own. This new

Anthropic launched a bold Super Bowl ad campaign mocking ads in AI chats. Their spots show funny interruptions, like fitness tips turning into insole pitches or therapy chats pushing dating sites. The tagline hits hard: “Ads are coming to AI. But not to Claude.” They backed it with a blog post pledging Claude stays ad-free

Sam Altman, OpenAI’s charismatic CEO, just dropped jaws in a Forbes interview. He revealed a wild succession plan: handing the company reins to an AI model someday. Altman argues if AGI can run businesses, OpenAI should lead by example. It’s his latest headline-grabbing move in a career full of big visions. He also claimed OpenAI