Mistral Launches ‘Shieldstral’, 3B Open-Weights AI Safety Model for Text and Images
Mistral AI has launched Shieldstral, a 3B open-weights safety model that moderates AI-generated text and images using plain-language rules.
Section
Category: AI & Technology
Latest AI, software, gadgets and technology news.
Liquid AI Releases ‘LFM2.5-2.6B’ for Efficient On-Device AI Agents
Liquid AI releases LFM2.5-2.6B, a 2.6B-parameter open-weight model built for efficient on-device AI agents.
Anthropic Signs $10 Billion Compute Deal With AI Cloud Startup Volta Infra
Anthropic AI news: a $10 billion, six-year compute deal with Nvidia-backed Volta Infra will power Claude from a data centre…
Razorpay Hires 4 Senior Engineering Leaders to Accelerate AI Strategy
Razorpay news: the fintech has hired four senior engineering leaders to build AI-first payments, fraud detection and business banking products.
Anthropic Brings Local Data Processing to Claude AI in India
Claude AI India now supports local data processing, letting enterprises keep eligible data inside the country for residency and compliance…
Apple Faces $32.5 Billion Lawsuit Over Photos App Face Scanning
An Apple lawsuit certified in Illinois seeks up to $32.5 billion, alleging the Photos app created biometric faceprints without users'…
EU AI Act Now Requires Labels on Deepfakes and AI Content
The EU AI Act's transparency rules took effect on August 2, 2026, requiring labels on AI-generated content and deepfakes, with…
Meta AI Uses a Second Agent as a ‘Memory Coach’ for Long Tasks
Meta's dual-agent design gives autonomous AI agents a dedicated memory coach that tracks failures and decides when a reminder is…
Only 4% of Government AI Projects Scale Beyond the Pilot Stage
Just 4% of government organizations scale AI into sovereign AI investment; 96% stall at pilots, blocked by data, governance and…
WhatsApp Tests Age Verification in India for Data Protection Law
WhatsApp is testing age verification for Indian users, asking some to confirm their date of birth under the DPDP Act.