Technology

Import AI 470: No rights for machines, SPADE environment generation, and Hawkeye GPU kernels

Import AI 470 covers METR's study on AI-accelerated vulnerability discovery, SPADE for automated environment generation, and Hawkeye for better GPU kernels.

Import AI 470 covers a METR study on AI-accelerated vulnerability discovery, SPADE for automating environment generation, and Hawkeye for optimizing GPU kernels. The newsletter continues Jack Clark's analysis of cutting-edge AI research and its implications.

Import AI 470, the weekly newsletter by Jack Clark (co-founder of Anthropic), covers three distinct technical developments reshaping AI infrastructure and governance [reference:31][reference:32].

No rights for machines

The lead story centers on a METR study finding that AI has dramatically accelerated cybersecurity vulnerability discovery in 2026, with CVE rates spiking for cURL, OpenSSL, and other critical infrastructure [reference:33]. The study raises questions about the governance implications of AI-accelerated vulnerability discovery and whether existing frameworks are equipped to handle the pace of change.

Automating environment generation with SPADE

The newsletter also covers SPADE, a system for automating environment generation. The technology promises to streamline the creation of training and testing environments for AI systems, potentially accelerating research and development cycles.

Building better GPU kernels with Hawkeye

Hawkeye is a tool for building better GPU kernels, addressing the performance challenges that arise as AI models grow larger and more complex. The development reflects ongoing efforts to optimize the hardware-software stack for AI workloads.

Context

Import AI is a weekly newsletter read by over 116,000 subscribers as of early 2026 [reference:34]. It covers cutting-edge AI research and its societal implications, with Clark's analysis providing context and connections between seemingly independent developments. Import AI 470 continues this tradition, drawing links between policy, technology, and research that together shape the trajectory of AI development.