AI News from February 12, 2026

Canadian grocer Loblaw partners with OpenAI to integrate PC Express delivery app into ChatGPT, enabling AI-powered menu planning and grocery ordering.

Australian retail conglomerate Wesfarmers expands Microsoft partnership, doubling Copilot footprint across Bunnings, Kmart, and other divisions for agentic commerce.

Chinese Academy of Sciences develops fruit fly-inspired compound eye with 1,027 visual units in 1.5mm sensor, enabling 180-degree vision and chemical detection.

Microsoft develops MAI-1-preview foundation model trained on 15,000 Nvidia H100 GPUs, reducing OpenAI reliance while maintaining partnership through 2032.

Google releases major upgrade to Gemini 3 Deep Think, achieving 48.4% on Humanity's Last Exam and gold medal performance on International Olympiad challenges.

Meta officially announces $10B+ investment in Lebanon, Indiana data center with 1GW capacity, creating 4,000 construction jobs and 300 permanent roles.

OpenAI unveils GPT-5.3-Codex-Spark, delivering over 1,000 tokens per second through partnership with Cerebras for ultra-low latency AI coding assistance.

AI startup Anthropic closes $30B Series G funding led by GIC and Coatue, achieving $380B valuation with $14B run-rate revenue growing 10x annually.

Programmers ask 9th Circuit to revive DMCA claims against OpenAI and Microsoft over AI trained on copyrighted code, seeking up to $9 billion in damages.

ASU integrates generative AI into first-year writing curriculum with expert colloquium series, targeting fall 2026 implementation across all campuses.

Stanford-founded Simile secures $100M led by Index Ventures with Fei-Fei Li and Andrej Karpathy investing in AI that predicts human decision-making.

CU partners with OpenAI for $2 million annual ChatGPT EDU deployment across four campuses, rolling out March 31, 2026 with privacy protections.

APT groups from China, North Korea, and Iran use Google Gemini for reconnaissance, malware coding, and phishing campaigns, Google GTIG reveals.

Zhipu AI unveils GLM-5, Moonshot launches Kimi 2.5, and ByteDance tests Seedance 2.0 as Chinese AI companies compete for the next breakthrough moment.

European Commission identifies key priorities for AI Act enforcement in 2026, focusing on general-purpose AI models and procedural rules.

Google DeepMind CEO forecasts new golden era of discovery with AI solving major challenges in medicine, energy, and space exploration.

Anthropic commits to paying for grid infrastructure costs and protecting consumers from electricity price hikes caused by its data centers.

OpenAI dissolves team responsible for communicating company mission to employees and public, reassigning members to other roles.

AI agents can now autonomously complete complex projects in hours, raising concerns about massive disruption to knowledge work and employment.

Gemini Deep Think achieves breakthrough performance in solving PhD-level math problems and enabling autonomous research in multiple fields.