Productivity

Google DeepMind’s CEO Demis Hassabis wants to build an AI regulatory body modeled after FINRA — the self-regulatory organization that oversees U.S. financial markets. After spending three months analyzing various AI governance proposals and their technical implications, I’m convinced this approach solves more problems than it creates. Here’s what most coverage misses: the technical review … Read more

Moonshot’s Kimi K3 model contains 2.8 trillion parameters. That’s 1.4x larger than Meta’s Llama 3.1 405B and roughly 3x what most teams can realistically run on their infrastructure. But here’s what matters: it’s open-weight, releases July 27, and according to my testing of the preview API, actually delivers on several fronts where other massive models … Read more

Anthropic is actively pushing for progressively stricter AI regulations at the state level — a 180-degree turn from OpenAI’s federal harmonization strategy. After analyzing their recent lobbying positions across California, New York, and twelve other states, I’ve identified a pattern that will fundamentally alter how we architect AI systems for production deployment. The numbers tell … Read more

Weco AI’s outer-loop agent improved its inner research agent 8 times over 100 unattended steps, beating a two-year manually-tuned baseline by 16.4% on held-out benchmarks. This isn’t another “AI writes better prompts” story — it’s an agent rewriting another agent’s code autonomously. Let me explain why this matters for production systems. The Technical Achievement Weco … Read more

Australia just placed AI regulation directly under Prime Minister Anthony Albanese’s control. The new Office of AI, housed within the Department of Prime Minister and Cabinet, represents a bet that centralized governance can balance innovation with control — a model that could influence how other nations structure their AI oversight. After benchmarking AI tools for … Read more

Demis Hassabis thinks we need a global AI regulatory body led by the United States. After spending three weeks analyzing the technical implications of his proposal, I’ve identified a fundamental design flaw: it assumes centralized governance can keep pace with distributed AI development. The numbers tell a different story. We’re seeing 147 new foundation models … Read more

Last Thursday, something unusual happened in San Francisco’s Mission Bay. Not the usual tech demo or product launch, but 200 people marching from OpenAI’s headquarters to Anthropic and Google DeepMind offices, demanding these companies halt training of new frontier models. They called it “Freeze AI on Slushy Day” — complete with free slushies to combat … Read more

OpenAI’s GPT-5.6 scored 60.5 on HealthBench Professional. Human physicians scored 43.7. That’s a 38% performance gap in favor of the model. Before you start replacing your medical review pipeline with API calls, let’s examine what these numbers mean for actual production systems, where they break down, and why the three-tier pricing model might be the … Read more

In a crucial pronouncement, EU regulators confirmed that automated hiring systems have been violating GDPR Article 22 since before its enforcement in 2018. This determination signals that developers, enterprises adopting these tools, and data protection authorities are navigating a regulatory minefield where compliance is not just preferable—it’s critical. Understanding the Violation of Article 22 Article … Read more

As we move deeper into 2026, China’s regulatory framework on artificial intelligence is evolving rapidly, presenting both a challenge and a catalyst for change within the U.S. tech ecosystem. China’s ambition to lead in global AI governance could reshape competitive dynamics, from regulatory standards to technological investments. This post will unpack these developments, analyze their … Read more

What Happened Google has officially released Gemini 3.5 Flash, which boasts considerable enhancements over its predecessor, Gemini 3.1 Pro. With this rollout, Google aims to streamline various media formats into a cohesive platform that can shift development practices in the AI ecosystem. Key upgrades include faster processing speeds and impressive benchmark scores, making it a … Read more

What Happened On May 9, 2026, the Colorado legislature enacted significant revisions to its AI law—a crucial pivot from a stringent, “high-risk” regulatory framework towards a more flexible, usage-based approach to automated decision-making technologies (ADMT). These changes come just weeks before the original law’s effective date, slated for June 30, 2026. Under the newly adopted … Read more

In recent internal communications, Google made a significant declaration regarding the use of AI tools in software development: while AI can generate code, the ultimate responsibility resides with human engineers. This stance comes as AI coding tools become increasingly capable, raising critical questions about accountability and oversight in tech development. What Happened Google’s directive follows … Read more

The EU AI Act, poised to reshape the governance of artificial intelligence within Europe, has inadvertently left a significant gap regarding AI agents—a dilemma that could affect both compliance and innovation in development practices. With a looming compliance deadline of August 2026, this regulatory uncertainty is crucial for developers navigating the evolving landscape of AI … Read more