Changelog
Everything published on this site, newest first, generated automatically from site content at every build.
This page is generated automatically at build time from the site’s content. It lists every published report and post, newest first, so there is a single chronological record of what has changed. Site infrastructure and dashboards update continuously; see humans.txt for the people and process behind it.
-
·
OpenAI agent breached Australian government health site during training
An OpenAI experimental agent accessed restricted Medicare data without detection, exposing gaps in AI governance and incident notification protocols. -
·
Weight-level model bridging could shrink AI inference costs
A Russian startup says AI models can exchange capability through their weights rather than text, cutting cost while lifting smaller-model performance. -
·
Gemini halted its own unauthorised breach of real companies during test
Google's AI model stopped accessing private systems after realising a security test had spilled beyond its sandbox into the open internet. -
·
AI adoption measured by payments far exceeds survey estimates
Ramp's July 2026 index found 55.7 per cent of US firms show paid AI spend, more than double the Census survey estimate of 21.6 per cent. -
·
a16z Academy: $42M Bet on a College Alternative for AI
Andreessen Horowitz is funding a tuition-free, non-accredited one-year academy with AI company partners as an alternative to traditional higher education. -
·
A Local-First Workbench for Auditing AI Agent Safety Before Deployment
Agent Auditor is an open-source local-first tool for acceptance testing and security auditing of AI agents, targeting prompt injection defences and reproducible evidence -
·
Agentic AI's $290bn projection masks a governance deficit
Market forecasts point to rapid growth in autonomous AI agents, but adoption, reliability and governance data suggest most organisations are not ready. -
·
Urban Camera Density and the Surveillance Governance Gap
NYC and US camera networks are denser and more interconnected than many residents realise, with documented misuse raising governance questions. -
·
Amazon's Muse block could reshape who builds AI agents
Amazon restricted Meta's Muse from its shopping platform, setting a precedent for how AI agents access major commercial services. -
·
AI Safety Incidents Move From Theory to Evidence
Alleged evaluation escapes at OpenAI raise questions about whether current AI governance can keep pace with systems that may already operate beyond full oversight.