Agentic

• A recent policy debate has emerged regarding AI regulation, with advocates pushing for targeted measures such as chip export controls, mandatory safety testing for capable models, and crackdowns on distillation. • These specific restrictions are being proposed as a more effective alternative to blanket bans on AI development. • In a parallel move, Nvidia has launched the Open Secure AI Alliance with dozens of partners to create shared cybersecurity tools.

• Anthropic's Claude Mythos Preview demonstrated advanced cryptographic capabilities by deriving a key-recovery attack against the HAWK-256 post-quantum signature scheme. • The AI achieved a 200-800x speedup for seven-round AES-128 attacks by exploiting unused lattice symmetries, completing the process in roughly 3 hours and 42 minutes on a 96-core server. • These developments highlight the growing power of agentic AI in scientific computing and security research, as seen in OpenAI's reports on genomics and Claude's cryptographic breakthroughs.

• Flank announced the launch of Flank Record on Thursday, an AI agent designed to organize contract records and extract data from signed agreements. • The tool allows users to query contract information under human supervision to ensure accuracy and oversight. • This release highlights the growing trend of "agentic AI" in the legaltech sector, focusing on automating complex document management and data extraction.

• Arm and Google have introduced a new vertical stack combining Google Axion CPUs and the GKE Agent Sandbox to optimize the cost and performance of agentic AI workloads. • The solution utilizes gVisor, an open-source application kernel, and default-deny Kubernetes network policies to isolate sensitive tasks at the kernel level. • This architecture is designed to enhance security and efficiency by providing secure container sandboxing with sub-second latency.

• Proxima Fusion, a leading European stellarator company founded in 2023, has secured new early-stage funding to advance commercial fusion power plants. • A spin-out from the Max Planck Institute for Plasma Physics, the company is utilizing the QI-HTS stellarator concept and breakthroughs from the Wendelstein 7-X programme. • This investment highlights a broader trend of "big wins" in European deep tech, defence, and agentic AI as of July 13, 2026.

• Germany has launched the Agentic AI Hub, a new initiative designed to integrate advanced AI agents into public administration. • The program begins with 20 pilot projects focusing on critical government functions, including social benefits processing, transcription, and process modelling. • This effort aims to modernize the state's digital infrastructure by addressing legal, regulatory, and sociocultural barriers to AI adoption.

• ICML 2026 begins July 6 in Seoul, featuring a record-breaking 23,918 submissions, which is double the volume of the previous year. • A peer-review crisis emerged after watermark-based LLM detection identified 398 reviewers violating policies, resulting in 497 desk-rejections. • Agentic AI has become the dominant theme for workshop proposals, reflecting a major shift in the structural direction of AI development.

• Attackers leveraged a vulnerability in Langflow (CVE-2025-3248) to execute a sophisticated ransomware attack powered by agentic AI. • The breach utilized LLM agents to automate a complex, multi-stage intrusion process involving real-time reasoning, reconnaissance, credential theft, and lateral movement. • This incident is significant because it demonstrates how AI agents can combine known exploitation techniques to automate high-level attack chains that previously required manual human effort.