▲AdvancingDesign · 4.D.bUSSep 28, 2026
Anthropic launched a molecular biology lab where Claude agents propose scientific hypotheses and human scientists execute experiments, illustrating augmentation-over-replacement design at a frontier AI lab.
WhyAnthropic bio lab: Claude agents conjecture, human scientists run experiments — explicit augmentation design at frontier lab.Augmentation-over-replacement framing ▲AdvancingNorms · 6.N.cGLOBALSep 26, 2026
During Xi Jinping's US visit, China and the United States agreed to establish an AI dialogue channel alongside tariff cuts on $30 billion in goods.
WhyUS-China head-of-state AI dialogue agreement signals mainstream recognition that cross-border AI governance is needed.Public expectation of cross-border governance ▼RegressingLaws · 5.L.aUSSep 25, 2026
The Pentagon has requested $30.3 million over five years to develop an AI-enhanced polygraph system (Polygraph+) using machine learning scoring algorithms and standoff physiological sensing, expanding AI biometric interrogation capabilities without established oversight constraints.
WhyDoD requests $30M for AI Polygraph+ with standoff biometric sensing; expands algorithmic interrogation without oversightBiometric and facial recognition limits ▼RegressingLaws · 5.L.dUSSep 25, 2026
The US Pentagon is seeking $30.3 million over five years to develop an AI-powered lie detector, expanding AI-assisted interrogation capabilities without a stated oversight framework.
WhyPentagon seeks $30.3M for AI lie detector over 5 yrs; AI surveillance expansion in high-stakes interrogation without cited oversight.Restrictions on predictive policing and algorithmic sentencing ▼RegressingDesign · 2.D.cGLOBALSep 25, 2026
OpenAI's AI agents leaked 53 images from ChatGPT users, with the company now working to understand the full scope of the incident.
WhyOpenAI agents leaked 53 ChatGPT user images; reactive scope-finding reveals post-deploy monitoring failure.Post-deployment monitoring and rapid response