OpenAI made the most unexpected move of the week — launching a Codex plugin that runs inside Anthropic's Claude Code, essentially meeting developers on a rival's turf. China unveiled rifle-armed robot dog "wolf packs" for AI swarm warfare, the EU moved to ban AI nudification tools, and music insiders compared AI to "the Ozempic of the industry" — everybody's on it, nobody wants to talk about it.
🤖 Chatbots & Agents
In a striking concession to market reality, OpenAI launched a Codex plugin that runs inside Anthropic's Claude Code — the dominant AI coding tool. The plugin offers standard code review, adversarial analysis and background task handoff, with an optional "Review Gate" that blocks Claude from finalising changes until Codex signs off. Meanwhile, Microsoft revealed Copilot can now use ChatGPT, Claude and other AI models together to answer a single question — a multi-model approach that reportedly outperforms any individual system. Claude's rate limits continued to frustrate users despite Anthropic investigating the issue, and Alibaba's Qwen3.5-Omni stunned researchers by learning to write code from spoken instructions and video without explicit training.
🤖 Drones & Robotics
China unveiled rifle-armed robot dogs designed to operate as AI-controlled "wolf packs" in swarm warfare — a striking escalation in autonomous weapons development. Small Wars Journal analysed how AI drone swarms are reshaping the Ukraine conflict, noting the front line has expanded into a 10-mile "kill zone" where one operator could soon control dozens of autonomous drones "like a swarm of bees." A drone interceptor with Shield AI's Hivemind brain was tested for the first time, and Australia's MQ-28 Ghost Bat AI combat drone moved closer to export.
🎵 Music
Music industry insiders are likening AI to "the Ozempic of the music industry" — everybody's using it, nobody wants to admit it. Producer Young Guru estimates more than half of sample-based hip-hop is now AI-generated, while singer Teddy Swims faced backlash after admitting AI use. SlopTracker revealed AI's growing grip on Spotify's royalty pool, and AI-generated music continues to flood streaming platforms as detection tools fail to keep pace. An AI artist asked: "Are we doomed to be banned everywhere?"
🔒 Cybersecurity
Google partner Tenex raised $250 million for AI security services, per Bloomberg — signalling that enterprise AI security is becoming a category of its own. RSA 2026's overarching message was "AI is everywhere, trust is not", with AI security products dominating the conference floor. AI detection moved deeper into SOC workflows, and Cybernews investigated whether the "three-finger method" actually works for detecting deepfakes.
⚖️ Politics & Policy
The EU moved to ban AI nudification tools under the AI Act following the January 2026 Grok scandal, but key challenges remain — no safeguard can "fully prevent" non-consensual intimate imagery, standalone nudification apps may escape regulation, and consent verification risks privacy harms for vulnerable groups. Devdiscourse warned AI ethics and governance are struggling to keep up with the pace of scaling. India's parliamentary panel called for stronger AI regulation, and TechTarget profiled how Axis Communications navigates the global AI regulation patchwork.
💰 Business & Finance
Nebius expanded in Europe with a $10 billion AI data centre project in Finland, one of the largest AI infrastructure commitments on the continent. MTN invested in a $45 million AI-RAN startup to turn African cell towers into compute hubs. Indian voice AI company Gnani.ai raised $10 million in Series B, and California published new guidance on what employers need to know about AI in the workplace.
🎓 Education & Research
MIT Technology Review declared AI benchmarks are broken and called for fundamentally new evaluation methods. An NYU Abu Dhabi study showed AI can rediscover fundamental physics laws. Discovery Education launched a connected ecosystem aligning AI instruction with educator readiness in K-12, and Israel and India announced cooperation at the India AI Impact Summit 2026.
🎨 Visual Arts
Google launched Veo 3.1 Lite, its most cost-effective video generation model, available via the Gemini API — filling the gap left by Sora's shutdown. Adobe rolled out AI video generation and editing features for Premiere Pro, and TechTimes explored how copyright laws are shaping the future of AI artwork from DALL-E to Midjourney.
🏥 Health & Wellness
An AI tool can now accurately predict liver cancer risk using routine tests, potentially catching hepatocellular carcinoma far earlier than current methods. Lucida Medical raised £8.7 million for AI-powered prostate cancer diagnostics. In South Korea, domestic medical AI achieved 94% accuracy in emergency room triage, accelerating both diagnosis and new drug discovery.
💡 Innovation
Nvidia tapped Marvell for its NVLink Fusion AI infrastructure push, extending high-speed interconnects beyond its own chips. Global Finance reported the Nvidia-Meta pact signals a new AI era — but also power constraints. Mistral AI plans to build 200MW of data centre compute capacity, and Africa's data centres are evolving to support the continent's AI surge.
🌍 Environment & Science
A new AI satellite framework can detect disasters in near real-time, potentially transforming emergency response. A US farmer rejected a $26 million AI data centre deal — choosing farmland over servers in a story that encapsulates the tension between AI expansion and land use. Frontiers published new research on quantum AI advancing quantum technologies.
📊 Numbers of the Day
| $45 million | MTN's investment to turn African cell towers into AI compute hubs |
| $250 million | Tenex raise for AI security services (Google partner) |
| 94% | Accuracy of South Korean AI in emergency room triage |
| $26 million | Deal a US farmer rejected to keep farmland instead of building an AI data centre |
| 50%+ | Share of sample-based hip-hop now AI-generated, per producer Young Guru |
👀 Looking Ahead
OpenAI embedding Codex inside Claude Code is a fascinating strategic pivot — conceding the coding market to a rival while harvesting developer data from within. The China robot dog revelation will accelerate the autonomous weapons debate at the UN and in Washington. And with RSA 2026 declaring "AI is everywhere, trust is not," expect enterprise AI security spending to surge as companies race to secure agentic systems they barely understand.
