AI License Plate Surveillance Cameras Throw a US Town Into a State of Emergency

According to The Washington Post, the deployment of AI-powered license plate reader cameras sparked a major controversy in one American city, ultimately culminating in a declaration of a state of emergency. It stands out as a textbook example of how increasingly sophisticated surveillance technology has inflamed debates over the balance between citizens' privacy and public safety.

While the rollout of license plate recognition cameras continues to spread across the United States, AI-driven advances have intensified concerns about privacy infringement. This case illustrates how the gap between the speed of technology adoption and public consensus-building can fuel social division. It is an incident that has once again called the state of AI surveillance technology into question.

German Intelligence Agency Turns Down Palantir, Picks a French AI Company

According to German outlet Politico, Germany's intelligence agency has chosen a French AI company rather than the US-based Palantir as the supplier for its AI analytics platform. The decision signals that European nations are beginning to take concrete action toward securing AI sovereignty.

Mistral's CEO recently warned that "Europe will become a vassal state of US AI unless it acts within two years," and Germany's decision fits squarely within that context. Even among NATO member states, moves to rethink dependence on American technology are spreading, and Europe's choices as a "third path" are drawing attention.

MiroThinker-1.7: An Open-Weights Deep Research Agent Built on the Qwen3 MoE Architecture

MiroThinker-1.7, an open-weights deep research agent, was announced in Reddit's LocalLLaMA community. It is built on Qwen3's MoE (Mixture of Experts) architecture, and the mini version sports a configuration of 30B total parameters and 3B active parameters, designed from the ground up to run on consumer-grade hardware.

The weights are publicly available on HuggingFace, and an API is also offered. It is a rare attempt at an open-weights deep research agent, and the community is showing strong interest in its inference speed on consumer hardware.

Gonfire: A New Tool for Evaluating AI Coding Agent Piloting Skills

On Hacker News's Show HN, Gonfire was released — a tool for evaluating how well developers pilot AI coding agents. It aims to shift away from traditional LeetCode-style technical interviews toward a format that assesses how skillfully candidates can wield AI coding agents.

According to developer, LeetCode had become firmly established as the hiring test for software engineers as of 2020, but recently there has been a marked shift toward case-study formats that encourage AI use in interviews for "AI engineer" roles. Take-home assignments and live sessions are emerging as the two dominant patterns, and a paradigm shift in technical interviews is well underway.

GitHub Copilot Moves to Weight-Based Billing Starting June 2026

According to an article on Qiita, GitHub Copilot's billing system will shift from the existing premium request model (count-based) to weight-based billing (token-based) on June 1, 2026. Voices like "the count-based system was easier to understand" are being raised, and concerns about costs are spreading through the developer community.

The article also rounds up cost-saving tips for the new billing system, making this a significant change with immediate impact for Japanese developers who use GitHub Copilot daily. High-usage teams should keep a close eye on fluctuations in their monthly costs.

AI-Polished Writing That Feels "Not Like Me" — The Struggle Students Face

An article in The Conversation reports on a phenomenon in which university students feel that their own AI-refined writing "sounds powerful, but isn't really me." The problem of AI "smoothing" of text flattening a writer's individuality is becoming increasingly visible.

Hacker News has also seen lively debate about how to read AI-assisted essays, with some saying "AI-generated text doesn't bother me as long as there's a minimal amount of effort put in," while others pose the question "how tolerant should we be of writing that smells of AI?" The meaning of "writing" in the AI age is being reexamined both in educational settings and in online communities.

---