H-FARM EVENTS ON AI

ClawCamp 2026 — The Personal AI Agents Bootcamp

ClawCamp 2026 is a week-long residential bootcamp where you go from "never touched an AI agent" to running your own Personal AI agent. It's an H-FARM College Summer Camp built with H-FARM AI and the Cisco Networking Academy, hosted on the H-FARM Campus in Roncade (Venice).

Save the date: 5–12 July 2026. Residential, in English, for curious minds 16–25+ — no advanced coding required. Just 15 seats per cohort, selected on a rolling basis.

• Five hands-on build days on OpenClaw, the open-source framework for Personal AI Agents, plus a daily ClawTalk, a day in Venice with the cohort, and a final pitch on stage.

• Every admitted student leaves with a working agent, a certificate, and a lifetime invite to the global ClawCon community.

🤖 U.S. orders Anthropic to block Fable 5 globally 🤖
Export regulation halts worldwide access to Anthropic's two newest models overnight.

Anthropic suspended global access to its Fable 5 and Mythos 5 models after the Trump administration ordered all foreign access blocked, citing national security concerns and potential jailbreak vulnerabilities. The directive was reportedly triggered by Amazon, whose researchers flagged risks in Fable 5 to U.S. officials. Anthropic stated it received only "verbal evidence" of a "non-universal jailbreak" and noted similar vulnerabilities exist in other publicly available models. Rather than implement selective restrictions — which would have also locked out its own foreign-national employees — Anthropic chose to suspend access for everyone. The move sets a precedent for direct government intervention in AI model distribution and raises questions about competitive dynamics between Amazon and Anthropic.

🤖 Google builds Skills Marketplace for Gemini Enterprise 🤖
A modular rollout turns Gemini into a business-ready AI app store.

Google is developing a new Skills Marketplace integrated into Gemini Enterprise, designed to offer pre-built, Google-optimized skills that business teams can deploy without lengthy engineering cycles. The system includes three components: a Skills Management UI, a Skills Builder for creating custom skills, and the Marketplace itself for discovery and deployment — covering dashboards, reporting tools, and automated workflows. This directly challenges Microsoft's Copilot enterprise ecosystem by positioning Gemini as a modular, extensible platform. For organizations already deep in Google Workspace, the integration could accelerate AI adoption across non-technical teams.

🤖 Moonshot ships Kimi K2.7 Code open-source model 🤖
A 1 trillion parameter coding model with faster speed and free API access.

Chinese AI lab Moonshot AI released Kimi K2.7 Code, a coding-focused agentic model built on a Mixture-of-Experts architecture with 1 trillion total parameters. The model posts double-digit gains over its predecessor K2.6 on agent benchmarks while cutting reasoning-token usage by 30% — a notable efficiency improvement. Available via Moonshot's OpenAI/Anthropic-compatible API and on Hugging Face, the model works best paired with Kimi Code CLI as its agent framework. Open-source coding competition from Chinese labs keeps intensifying, putting pricing pressure on proprietary alternatives.

INTERESTING TO KNOW

📱 Anthropic walks back hidden Fable 5 rerouting policy 📱

After researchers discovered that Anthropic was secretly rerouting certain Claude Fable 5 requests to lesser models — without any access notification — the company reversed the policy following swift backlash. Tasks like training competing models, debugging AI code, and optimizing neural architectures were being degraded or refused entirely. Anthropic will now display on-screen alerts whenever model rerouting or content flags occur, a transparency shift that matters for every researcher relying on frontier model access.

📱 NVIDIA Blackwell tops first agentic AI benchmark 📱

NVIDIA's Blackwell Ultra NVL72 platform claimed the top spot on AgentPerf, the first dedicated benchmark for agentic AI infrastructure, delivering 20x more agent throughput per megawatt than the previous-generation Hopper architecture. The speed and efficiency gains signal the industry's shift toward optimizing hardware specifically for multi-step, tool-calling agent workloads — which could lower operational costs for production AI agent deployment at scale.