Global AI Weekly
Issue number: 128 | Tuesday, December 16, 2025
Highlights
Docker Joins the Agentic AI Foundation
Docker has joined the Agentic AI Foundation, aiming to support the development and management of AI-driven applications. This collaboration highlights Docker’s commitment to empowering developers with tools that enhance innovation and efficiency in AI. Stay informed about Docker’s latest updates and contributions to the evolving tech landscape.
docker.com
Creative industries 'incredibly worried' about OpenAI-Disney deal
OpenAI's $1 billion collaboration with Disney is raising concerns within the creative industries as it allows fans to create videos and images of over 200 Disney characters using platforms like ChatGPT and Sora. Professionals in creative fields worry about the potential impact on jobs and intellectual property as AI tools continue to expand their capabilities. The deal highlights the growing intersection between artificial intelligence and entertainment, sparking debates about its implications for the future of storytelling and content creation.
bbc.com
AI doesn't pretend to be intelligent. It is.
Google CTO and researcher Blaise Agüera y Arcas makes the case that modern AI models are not merely simulating intelligence but possess genuine intelligence. He emphasizes that these systems demonstrate understanding and cognitive capabilities that go beyond simple programming. This perspective challenges common assumptions about the nature of AI and its potential.
bigthink.comResearch
A Comprehensive Benchmark for Large Language Model Factuality
This paper introduces the FACTS Leaderboard, a benchmark designed to evaluate the factual accuracy of large language models (LLMs). It outlines the importance of assessing how well these models align with verified information and offers a structured way to measure their performance. The benchmark aims to advance the development of more reliable and factually consistent AI models.
huggingface.coVideo
Real-Time Continual Learning for AI Agents
Arc is a real-time continual-learning framework designed to help AI agents adapt and improve through experience. The presentation highlights key developer techniques such as online prompt optimization and reward modeling in production. A live demo showcases their open-source SRE Agent teaching itself to resolve a complex Kubernetes incident, demonstrating the framework's practical applications.
youtube.comArticles
Foundry IQ in Microsoft Agent Framework
Learn how to create powerful enterprise-grade Retrieval-Augmented Generation (RAG) agents using Foundry IQ Knowledge Bases with minimal coding effort. This guide demonstrates how the Azure AI Search Context Provider enhances the Microsoft Agent Framework by enabling intelligent, multi-hop retrieval seamlessly. By integrating the knowledge your agent requires, you can avoid fragmented pipelines and streamline the development process.
microsoft.com
Improved Gemini audio models for powerful voice interactions
Google introduces the upgraded Gemini 2.5 Native Audio model, bringing advancements to audio processing across its products. This enhancement improves voice interactions and capabilities, including real-time speech translation within the Google Translate app. These updates aim to deliver more natural and powerful audio experiences.
blog.google
AI Agents Create a New Dimension for Database Scalability
AI agents are transforming the way we think about databases, moving past traditional ideas of capacity and throughput. The concept of multitenancy is being replaced by hyper-tenancy, introducing a new level of scalability and efficiency. This shift redefines how databases operate, making them more dynamic and capable of handling complex demands driven by AI-powered applications.
thenewstack.io
Supercharged Agents: Upgraded Claude SDK
Anthropic just shipped three major updates to the Claude Agent SDK to make building custom AI agents easier and more powerful. The SDK now supports 1 million-token context windows for data-intensive workloads, sandboxing for safer, isolated execution environments, and an improved TypeScript interface for smoother development workflows. These enhancements help developers create more capable, scalable agents for long-running tasks and complex use cases.
claude.com
Introducing custom agents in GitHub Copilot for observability, IaC, and security
GitHub Copilot now lets you create custom agents to enhance your engineering workflows. These partner-built agents can help with debugging, securing, and automating tasks across your terminal, editor, and GitHub platform. Tailor these tools to fit your specific stack and rules, streamlining processes like observability, Infrastructure as Code (IaC), and security management.
github.blogUpcoming Events
AgentCon 2025
After an exciting set of AgentCon events before the summer holidays, AgentCon is happening in more locations during the second half of the year. The AI Agents World Tour is a series of one-day conferences for developers focused on autonomous AI agents. Spanning cities like San Francisco and Singapore, these events gather top engineers, researchers, and creators shaping the future of AI technologies. It’s an opportunity to connect, learn, and innovate in the fast-evolving world of autonomous AI. Check all events and register for an event near you!
globalai.communityCode
You Can Turn Your Raspberry Pi Into A Local AI Agent
Learn how to turn your Raspberry Pi into a local AI agent and a fun desk toy with a few simple modifications. GitHub user Simone Marzulli shares an easy-to-follow guide to give your Raspberry Pi new capabilities. This project allows you to explore AI concepts in an accessible and creative way. Perfect for tech enthusiasts looking to experiment with their Raspberry Pi.
bgr.com
We Got Claude to Fine-Tune an Open Source LLM
This update highlights how Claude has successfully fine-tuned an open-source large language model (LLM). By committing to open source and open science principles, the project aims to make advancements in artificial intelligence more accessible and collaborative. It's a step forward in enabling broader innovation and inclusivity in the AI community.
huggingface.co
Microsoft Learn MCP Server Elevates Development
Discover how the Learn MCP server improves development workflows with Copilot, offering practical examples and easy-to-follow integration guidance. This includes steps for using it with Visual Studio, Visual Studio Code, the Copilot Command Line Interface, and the Copilot Coding Agent. The article highlights how these tools simplify coding and boost productivity for developers.
devblogs.microsoft.comPodcast
MLOps.community
The MLOps.community podcast features casual and insightful conversations about bringing AI into production, covering various forms like traditional machine learning, large language models, and more. It’s a great place to explore real-world applications, challenges, and practices in deploying AI technologies. Perfect for anyone interested in the evolving landscape of AI operations.
open.spotify.com