H-FARM EVENTS ON AI

ClawCon Venice — The Personal AI Festival

ClawCon is the gathering for the Internet's most curious personal AI community and it's coming to the H-FARM Campus. What started as a niche OpenClaw meetup quickly turned into a global movement: builders, researchers, designers, and curious humans meeting IRL to test what happens when AI becomes open and social-first.

ClawCon Venice is Italy's first ClawCon, hosted right here on the H-FARM Campus. Save the date: Thursday 4 June, 18:00–21:00. Here's what you'll find:

• A global community of OpenClaw builders, researchers, and designers meeting IRL, plus live demos of open, social-first AI projects.

• Unfiltered, hands-on conversations on where personal and agentic AI is heading next, in the OpenClaw spirit: curious, open, and refreshingly social.

Whether you build, research, design, or are simply AI-curious: there's a seat for you.

🤖 Demis Hassabis predicts AGI by 2030 🤖
DeepMind's CEO outlines AGI's rollout timeline and remaining capability gaps.

In a sit-down interview at Google I/O, Demis Hassabis shared his conviction that AGI will arrive by 2030, give or take a year. The Google DeepMind CEO identified the remaining gaps — world physics, memory, consistency, and continual learning — while expressing growing confidence the field is on track. He pinpointed oncology and immunology as the first diseases AI will crack and argued humans will retain an edge in taste, original thinking, and emotional connection. Post-AGI, Hassabis plans to pivot toward studying the nature of reality using AI. His timeline signals how the industry's most powerful lab is allocating resources, shaping competitive pressure across the entire field.

🧠 Claude Mythos cracks Erdős Problem #90 🧠
Anthropic's frontier model matches OpenAI's speed in deep math reasoning.

Anthropic's Claude Mythos has reportedly solved Erdős Problem #90, the same landmark mathematical challenge that OpenAI recently cracked, arriving at the result with what observers called a "cute, simple proof." While Mythos' initial solution was slightly less optimal, the model was also able to reproduce OpenAI's stronger proof. The result shows frontier mathematical reasoning is no longer exclusive to a single lab — a clear sign that deep-reasoning capabilities are commoditizing fast. Anthropic's latest models are operating at a comparable level in formal reasoning, intensifying the model-capability race between the two leading labs.

⚡ China curbs AI travel, xAI-Cursor deal scrutinized ⚡
Regulation, integration risks, and new benchmarks reshape the AI landscape this week.

Three notable developments highlight different pressures on the AI industry. First, xAI's top lawyer warned employees to carefully moderate interactions with Cursor staff — a standard gun-jumping precaution as the two teams collaborate ahead of an acquisition; any improper co-mingling could jeopardize the deal. Second, DeepSWE introduces a contamination-free benchmark for long-horizon software engineering spanning 91 repositories in five languages, providing sharper differentiation among coding agents. Third, China has expanded overseas travel restrictions to top AI talent at private firms, including executives at DeepSeek and Alibaba — tightening geopolitical control over the country's most valuable technical minds and raising regulation risks for global AI collaboration.

INTERESTING TO KNOW

🎨 Microsoft MAI-Image-2.5 debuts #3 on Arena 🎨

Microsoft's MAI-Image-2.5 has rolled out at third place on Arena's text-to-image leaderboard, excelling in style variety, accurate text rendering, and detailed imagery. The model represents a major upgrade over MAI-Image-2, with improved visual reasoning and commercial illustration capabilities — creating real competitive pressure in image generation alongside established players. Expect broader integration across Office, Designer, and Azure products soon.

🔧 NVIDIA CompileIQ ships 15% GPU speed boost 🔧

NVIDIA has shipped CompileIQ, integrated into CUDA 13.3, which uses AI-driven evolutionary algorithms to auto-tune GPU compiler settings per individual kernel, delivering up to 15% speed gains on already-optimized AI inference and training tasks. Developers can define multi-objective optimization goals, balancing runtime, power consumption, and compile time — especially valuable for latency-sensitive LLM inference at scale. The tool deepens NVIDIA's software ecosystem moat alongside its hardware dominance.