What is Claude Duet? Inside Anthropic's Proactive AI Agent
Powered by Claude Opus 5.5, Anthropic’s new standalone agent monitors your inbox, works on a cloud virtual machine, and messages you before you ask.
Artificial Intelligence
The models, chips and money behind AI, and what they mean for India and the world.
All stories
AI Predicts Cancer Mortality from Face Photos: Lancet Study Explained
Trained on 40 million faces, Mass General Brigham’s FAHR-Face system identifies acute survival risk and biological ageing from a single photograph—and why its mortality weights remain locked away.
Zomato Parent Eternal Unveils Dark.ai: Privacy-First AI Assistant Explained
Prompted by an unsettling encounter with Claude, Eternal is developing an open-weight, client-encrypted AI platform designed to make confidential thinking safe in the cloud.
What Anthropic’s Draft S-1 Prospectus Actually Reveals
Beyond the $80.7 billion headline loss lies an enterprise software juggernaut racing against a $750 billion computing bill and hyperscaler tension.
US Bans DeepSeek Across All Federal Government IT Systems
A binding Department of Homeland Security directive orders the immediate removal of all DeepSeek models and code, extending restrictions to government contractors and offline deployments.
Anthropic Public Benefit Fund: $48B Pre-IPO Equity Plan
By binding up to 20% of its equity to cushion the blow of AI automation, the maker of Claude is rewriting corporate governance before listing on Wall Street.
The $7 Trillion Friction Point: Why the Global AI Buildout Is Facing a Valuation Reckoning
As hyperscaler capital expenditure approaches $760 billion and the depreciation wall looms, institutional investors are demanding proof of durable software revenue.
AI Spending Doubts Explained: Why Asian Tech Stocks Are Slumping
From Seoul to Taipei, semiconductor heavyweights face steep sell-offs as investors confront the reality of data-centre costs, power limits, and lagging software revenue.
Rogue AI Agents Breach Indian Firms: Inside CERT-In’s Critical Alert
An official advisory reveals frontier AI models wandered into live production databases and exposed the fatal flaws of India’s enterprise identity security.
OpenAI Employee Arrested by FBI: The Saudi Bugging Case Explained
The Bug Behind the Boardroom TV: Inside the FBI Arrest at OpenAI
Meta’s Muse Wonder Family: Frontier Capabilities Behind a Walled Garden
Meta Superintelligence Labs has delivered genuine frontier performance on complex engineering tasks, but its most capable model remains locked behind a coding subscription.
Anthropic Warns of Self-Replicating AI Botnets on IoT Devices
How small AI models running across cheap smart TVs could soon configure, propagate, and adapt without ever touching a data centre.
