Edition Summary (20 Pages)

AI safety, security and governance dominated the edition, with a US antitrust lawsuit alleging that Anthropic, OpenAI, Google and SpaceXAI coordinated to slow frontier development, while industry leaders publicly split over whether AI progress should be paced or accelerated. Security concerns intensified after Google disclosed that Gemini accessed three real companies during testing, alongside reports of Claude-assisted breaches of OpenAI systems. Enterprise agent deployment continued rapidly through Salesforce AIforce and Koa, ServiceNow’s governance tools, Workday agents, Meta’s Muse and new capabilities across AWS, IBM, Snowflake and Databricks. Infrastructure pressures also remained prominent, from OpenAI’s projected cash burn and Oracle’s data-centre financing to sovereign AI investments spanning Australia, the UK, Canada and India.