Weekly AI News Roundup: Microsoft, Anthropic, OpenAI, Apple, Google, Meta Updates

AI in Enterprise & Productivity

• 💡 Microsoft 365 Copilot trials with 20,000 UK government employees showed an average saving of 26 minutes per day per worker, equivalent to two weeks annually.
• 🚀 OpenAI's Sora is now available for free on Bing, allowing users to generate 5-second vertical AI videos, with landscape support coming soon.
• 🧠 OpenAI Codex is now accessible to ChatGPT Plus users, offering internet connectivity and enhanced capabilities for coding tasks.
• ✅ Google NotebookLM introduced public sharing, enabling collaborative creation of study guides, product hubs, and research findings.

Major AI Model & Platform Updates

• 🛡️ Anthropic launched Claude Gov, a specialized AI model designed for US defense and intelligence agencies, deployed in secure environments.
• ✍️ Anthropic Explains is a new blog primarily written by Claude, demonstrating the model's ability to generate complex content with human oversight.
• 🍎 Apple is testing a 150 billion parameter LLM internally, aiming to approach the quality of recent ChatGPT rollouts, with Google Gemini and Perplexity potentially becoming Siri backend alternatives.
• ⚠️ Anthropic cut Windsurf's direct access to its models amid rumors of OpenAI acquiring Windsurf, citing a focus on sustainable partnerships.

Disruptive AI Applications

• 🎯 Perplexity's CEO predicts AI agents will redefine web interaction, moving beyond answering questions to taking actions like booking and ordering.
• 📈 Meta aims to automate ad creation entirely by the end of 2026, allowing businesses to set objectives and connect bank accounts for automated, personalized ad campaigns.
• 📞 Phone.com achieved 99.2% conversational accuracy with its AI phone agents, significantly reducing response times through a partnership with Groq's inference platform.
• 🎭 Humen launched AI Studio, enhancing human-like avatars with features like voice director for fine-tuned speech and gesture control, with future capabilities including prompt-based camera motion.

AI's Impact on Health & Science

• 🩺 The FDA approved Clarity Breast, an AI tool that predicts a five-year risk for breast cancer using standard mammograms, marking a significant milestone for AI in public health.
• 🌌 New AI techniques are improving the analysis of cosmological parameters, allowing researchers to identify subtle patterns in vast datasets and increase prediction accuracy by 30%.

Industry Shifts & Corporate Battles

• 🔄 Microsoft's AI team is undergoing its third major restructuring since early 2024, with significant leadership changes to integrate AI across its products.
• ⚖️ Reddit filed a lawsuit against Anthropic for allegedly breaching contract and attempting to use user data without consent, highlighting ongoing battles over data licensing.
• 📉 Apple faces significant innovation challenges, with concerns raised about its slow AI progress and potential market share threats from new partnerships like OpenAI and Jony Ive.
• 🧩 Meta is restructuring its generative AI group into two units, focusing on foundational models (Llama) and AI integration into consumer apps like WhatsApp and Instagram.