ENFR
8news

Tech • IA • Crypto

TodayMy briefingVideosTop articles 24hArchivesFavoritesMy topics

UK Forces Google AI Controls; ChatGPT Agents, Claude Fable 5 Launch

AITuesday, June 9, 2026· 25 videos

Briefing

Audio player
0:00 / 0:00

UK forces Google AI content controls

United Kingdom regulators will require Google to give publishers granular control over how content is used in AI systems. Website owners can block usage for AI training or grounding, including at the page level. New Search Console tools will show how content appears in AI Overviews, increasing transparency. The move could reshape search economics and set a precedent across Europe.

ChatGPT Workspace Agents target enterprises

OpenAI introduced ChatGPT Workspace Agents to automate multi-step workflows across teams. Powered by GPT-5.5, the agents can gather data, reason across systems, and execute tasks using tools like email and calendars. A no-code builder allows business users to create custom agents with natural language. The launch aims to fill the gap between individual AI tools and full enterprise system overhauls.

OpenAI pushes ChatGPT super app vision

ChatGPT is being redesigned into a full AI agent platform capable of executing tasks autonomously. Deep integration of Codex enables control over software, workflows, and external tools. Early use cases include managing emails, scheduling, and generating daily briefings. The long-term goal is an “invisible interface” where users assign goals and AI handles execution.

Anthropic launches Claude Fable 5

Anthropic released Claude Fable 5, its most capable public model derived from the restricted Mythos 5 system. The model shows major gains in reasoning, coding, and vision but includes strict safeguards for sensitive domains. Access to Mythos 5 remains limited under Project Glass Wing with government-linked partners. Pricing and safety constraints highlight rising tension between capability and risk.

Apple unveils AI-first Siri overhaul

Apple used WWDC 2026 to showcase a rebuilt Siri with deep system-wide AI integration. The assistant now supports contextual, conversational interactions across iPhone, Mac, and Vision Pro. A privacy-focused “private cloud” architecture blends on-device and secure cloud processing. الأسواق reacted cautiously, with volatility reflecting doubts about Apple’s competitive position.

US considers stakes in OpenAI

U.S. policymakers, including Donald Trump, are exploring government equity stakes in firms like OpenAI and Anthropic. Proposals suggest companies could grant shares to the state, creating a hybrid public-private model. Advocates like Bernie Sanders argue AI wealth should fund national programs. The idea signals a major shift in how AI is treated as strategic infrastructure.

NSA reportedly uses Anthropic Mythos

Reports claim the NSA may be deploying Anthropic’s Mythos model for offensive cyber operations. The system excels at identifying software vulnerabilities, offering potential strategic advantages. This use appears to conflict with Anthropic’s stated safety commitments. The development raises new concerns about AI’s role in cyber warfare and global power dynamics.

Sakana AI launches RSI Lab

Sakana AI introduced its RSI Lab, focused on recursive self-improvement systems. The approach aims to build powerful AI without massive compute by iteratively refining models. Inspired by evolutionary processes, systems prioritize promising solutions over brute-force scaling. This strategy challenges the dominant paradigm of compute-heavy AI development led by U.S. tech giants.

Videos covered

Previous briefings · AI