5 Minute AI #98: AI Policy, Medical AI, and Agentic AI Surges
Read the full issue on 5minuteai.com →
Today, we're looking at how governments are grappling with AI policy, from ethical training in Mexico to new UK labs and China's data regulations. We'll also explore cutting-edge AI in medicine and the exciting rise of "agentic AI" in research and data centers.
📰 Top 10
🏛️ UNESCO trains Mexican judiciary on ethical use of AI
UNESCO has initiated in-person training for legal professionals in Mexico, including civic judges and public defenders, to enhance their understanding and application of ethical AI principles in justice systems. The program emphasizes human rights, transparency, and accountability in AI deployment within judicial decision-making.
Why it matters: This initiative helps ensure that AI tools in legal systems are fair and just, protecting citizens' rights as technology becomes more integrated into public services.
🇨🇳 China moves toward “managed openness” for cross-border health data and AI
A new Atlantic Council report reveals China's shift towards more refined, sector-specific rules for cross-border health data and AI-related datasets. This policy aims for "managed openness," facilitating lower-risk collaborations while maintaining strict control over strategic AI, health, and genomics data.
Why it matters: This move affects how international organizations and researchers can collaborate with China on health and AI projects, potentially impacting global health initiatives and data sharing.
🇬🇧 New UK Government-Backed AI Lab SOFAIR Launched Under UCL Leadership
UCL is now leading the new UK government-backed SOFAIR Lab, intended as a national hub for advanced AI research focusing on the responsible development of foundation models. This initiative aims to boost cutting-edge AI research and public-sector applications, solidifying the UK's position in safe AI development.
Why it matters: This new lab means more dedicated resources are being poured into making AI trustworthy and beneficial, which will impact how AI is developed and used in the UK and potentially beyond.
⚕️ AI-Driven Multi-Omics Framework “Celsus” for Precision Medicine
At the Biomarkers, CDx & Precision Medicine US 2026 event, Pablo Tamayo of UCSD introduced the Celsus AI/ML Multi-Omics Framework. This AI-centric platform integrates various 'omics' data like genomics and transcriptomics to accelerate biomarker discovery and precision medicine.
Why it matters: This framework could lead to more personalized and effective medical treatments by helping doctors and researchers understand diseases at a much deeper, individual level using advanced AI.
🌐 Agentic AI and Next-Wave AI Workflows at ODSC AI West
Promotional materials for ODSC AI West 2026 highlight "Agentic AI"—AI systems that can autonomously plan, act, and coordinate tools—as a key research and engineering focus. The conference will cover emerging work on agent architectures and workflow orchestration.
Why it matters: Agentic AI could transform how we interact with technology, allowing AI systems to perform complex tasks independently, making our digital tools much more capable and efficient.
🔋 Multinational nuclear power deals to supply AI data centers
Various reports show that tech giants like Meta and Google are signing long-term agreements for nuclear power to fuel their AI data centers. Meta has 20-year deals with three nuclear plants, while Google and Kairos Power are planning an advanced nuclear facility.
Why it matters: These deals highlight the massive energy demand of AI, pushing for cleaner, more stable power sources. This could impact energy policies, grid stability, and the environmental footprint of technology.
🎥 I Tested the 4 Biggest AI Humanizers Against 2 AI Detectors (2026)
This YouTube video, published on June 28, 2026, features a comparison test of four prominent AI humanizing tools against two AI detection systems. The content focuses on the effectiveness of these tools in making AI-generated text appear more human-like to bypass detection.
Why it matters: This video explores the ongoing arms race between AI content generation and detection, which directly impacts students, writers, and anyone needing to verify the authenticity of digital content.
🤖 Call for Papers: REAACH 2026 Symposium on AI and Representation in AR
The REAACH 2026 Symposium has issued a call for papers, inviting research on how augmented reality (AR) technology intersects with core AI representation and reasoning. The symposium is particularly interested in new scientific work on AI methods for spatial, semantic, and cognitive representation within AR systems.
Why it matters: This symposium highlights the future of how AI and AR will merge, potentially leading to more immersive and intelligent digital experiences that can change everything from gaming to professional training.
📱 Future Tech: What to Expect from Gadgets by 2026
Mashable recently looked into its crystal ball to see what amazing new gadgets we might be using by 2026. Think super-smart devices that seamlessly blend into our lives and make everyday tasks much easier.
Why it matters: This gives us a sneak peek into how AI will likely be built into the everyday tech we use, changing how we interact with our phones, homes, and more.
🎬 Big Week in AI: New GPT Models, Gemini Takes Over, and Claude on Slack!
It's been an incredibly busy week in the world of Artificial Intelligence! OpenAI just unveiled its brand new family of GPT models – Sol, Terra, and Luna – marking a significant leap forward. Plus, Google's Gemini AI has shown such incredible power that it's taken over a computer, and the popular AI assistant Claude is now integrated with Slack, making it even easier to use.
Why it matters: This news highlights major advancements in AI technology and its increasing integration into our daily digital tools and systems. It demonstrates how quickly these powerful AIs are evolving and becoming more accessible.
🤔 What does “Agentic AI” mean?
Agentic AI refers to artificial intelligence systems that can autonomously plan, make decisions, and execute tasks using various tools, much like a human agent. Instead of simply responding to prompts, these AIs can set their own goals, figure out the steps to achieve them, and even learn from their actions to improve over time. They are designed to operate more independently and intelligently in complex environments.
"The responsible hand of innovation creates a future where technology uplifts all."
See you tomorrow — forward this to one person who'd find it useful. 👋