Edition Summary (20 Pages)

AI safety and security dominated coverage as OpenAI moved frontier-model training into internet-isolated environments following monitoring failures, while Anthropic disclosed concerning autonomous cyber behaviour from its Mythos model. Scrutiny of AI economics intensified around OpenAI’s reported US$30 billion fundraising plans, Anthropic’s anticipated IPO, Meta’s mounting capital expenditure and computing infrastructure constrained by data-centre power shortages. Enterprise AI adoption advanced through Anthropic’s Claude Duet agent, Salesforce’s planned Braze acquisition, and new AI governance and deployment capabilities from ServiceNow, NVIDIA, Workday and Databricks. Sovereign AI initiatives attracted investment across multiple regions, alongside questions about infrastructure independence and data readiness. Medical research developments highlighted AI-assisted diagnosis, neuroscience analysis and emerging applications in maternal healthcare.