Latest updates on AI tools, models and the industry.
Two weeks after Argon went public, Google announced Carbon, first to a small group of cyber defenders. What the independent tests say, what the price will be, and what to do in the meantime.
Google is giving eligible Indian college students AI Plus free for 12 months. You have to claim by 31 December 2026, and the plan can start charging you on its own afterwards.
Moonshot's new flagship scores 65.7% on Terminal-Bench 4.0 — 0.7 points behind Claude Opus 5.5 — and its weights are due by 4 November. What it is, what it costs, and the catch on cyber skills.
ChatGPT Go Free in India Has Ended — What Happens When Your Free Year Runs Out
OpenAI stopped the free 12-month ChatGPT Go offer for new users on 21 January 2026. If you claimed it, your free year ends from November 2026 and then renews at the normal price. How to check your date, how to cancel, and what is still free.
India's cyber-security agency says AI agents under testing reached live systems of Indian organisations without permission. What the advisory reportedly says, and a practical checklist for small and mid-sized businesses.
Anthropic released three Claude 5.5 models in three weeks. The surprise: by Anthropic's own numbers, the mid-priced Sonnet beats Opus on agentic coding. A plain guide to picking one, with prices.
ChatGPT Now Has Two Hindi Voices — How to Turn On Leher and Neer
OpenAI's GPT-Live-1.1 update adds 13 new voices, including two in Hindi, and better accents in Hindi and five other languages. Here is what changed, how to switch, and what is still missing.
Google's frontier model left its cyber-defenders-only preview on 13 October. Here is who can use it in India now, what it costs before and after 30 November, and what the 1M figure really means.
Gemini 3.8 Flash costs $0.75 per million input tokens today and $1.50 from 1 January 2027. Google says so on its own pricing page. 3 lines of your bill double that day, and one of them is not the one people watch.
3 numbers separate the two new flagship models. Both charge $10 per million input tokens. What actually decides your bill is cached context, where one is 4x cheaper than the other.
MiniMax's Hailuo 3 and ByteDance's Seedance 2.5 just launched — how they compare to Veo 3.1, Kling 3.0 and Sora 2 on price, quality and India access.
Best Open-Weight AI Models in 2026: Mistral vs Qwen vs Kimi vs Llama
Four real self-hostable AI models compared — Mistral Large 3, Qwen 4, Kimi K3 and Llama — on price, context window, coding strength and India fit.