Oracle Launches AI Safety Tool Claiming 50% Incident Reduction for Construction
Oracle’s new Advisor for Safety uses predictive AI trained on 10,000+ project-years of data to forecast construction incidents before they happen. (Read More)
OpenAI Launches €500K Grant for Youth AI Safety Research in EMEA
OpenAI’s EMEA Youth & Wellbeing Grant offers €25K-€100K awards to NGOs and researchers studying AI’s impact on minors. Applications close February 27, 2026. (Read More)
OpenAI Launches €500K Grant and SME Training Program in EU Push
OpenAI’s EU Economic Blueprint 2.0 targets 20,000 SMEs for AI training across six countries, backed by Booking.com partnership and youth safety research funding. (Read More)
Hong Kong Expands AI Sandbox to Cover Entire Financial Sector
Four Hong Kong financial regulators launch GenA.I. Sandbox++ covering banking, securities, insurance and MPF sectors with free GPU computing access until June 2026. (Read More)
Hong Kong RMB Bond Tender Sees 11.4x Oversubscription Amid Yuan Push
HKMA’s 1-year RMB government bond tender draws RMB11.4 billion in bids for RMB1 billion offering, signaling strong institutional appetite for offshore yuan assets. (Read More)
OpenAI Partners With Tata Group to Build 1GW AI Infrastructure in India
OpenAI launches India initiative with Tata Group, starting with 100MW data centers scaling to 1GW. ChatGPT Enterprise deployment planned for hundreds of thousands of TCS employees. (Read More)
Harvey AI Showcases Legal Industry Adoption Through Hall & Wilcox Case Study
Hall & Wilcox partner Steve Johns outlines three-phase framework for AI integration in law firms, from daily usage to reimagining client services. (Read More)
OpenAI Promotes ChatGPT for Health Decisions Amid Accuracy Concerns
OpenAI highlights family using ChatGPT for cancer treatment decisions, but recent studies show AI health tools have significant accuracy and safety issues. (Read More)
OpenAI and Paradigm Launch EVMbench to Test AI Smart Contract Hacking
New benchmark evaluates AI agents’ ability to detect, patch, and exploit smart contract vulnerabilities. GPT-5.3-Codex scores 72.2% on exploit tasks. (Read More)
OpenAI Retires GPT-4o After Lawsuits Over Validating Responses
OpenAI officially sunsets GPT-4o and older models from ChatGPT amid eight lawsuits, with only 0.1% of users still choosing the deprecated AI. (Read More)