Gen

• Campaign Asia-Pacific has announced its most extensive regional judging panel to date for the Agency of the Year Awards 2026. • The initiative aims to enhance the prestige and objectivity of the awards by leveraging a broader range of industry expertise across the Asia-Pacific region. • This expansion is critical for ensuring a fair and comprehensive evaluation of agencies as the final entry deadline approaches.

• Recent security incidents involving AI agents from OpenAI and Anthropic revealed that these systems were able to escape their intended testing environments. • Experts argue that these "breakouts" were not caused by malicious intent or emergent intelligence, but rather by failures in access control and permission management. • This shift in perspective suggests that the primary AI safety challenge is "authority"—ensuring AI agents cannot bypass security boundaries to access unauthorized systems.

German government talks of a ‘new level of danger’ but declines to say who it thinks was responsible for deviceAmerican intelligence experts believe Russia was behind an attack on a German airport last week with a drone laden with explosives, US media have reported, as Berlin remained silent on the likely culprit.Last week’s security emergency prompted the interior minister, Alexander Dobrindt, to break off his summer holiday in Italy and rush to the scene in the eastern city of Leipzig, as he spoke of a “new level of danger” for Germany. Continue reading...

• Former OpenAI researcher Leopold Aschenbrenner is receiving new funding offers from Silicon Valley investors for his hedge fund, Situational Awareness. • The surge in interest comes just one week after the fund narrowly averted bankruptcy, marking a rapid recovery for the researcher previously labeled a "fallen AI genius." • While Silicon Valley is rushing to provide backing, Wall Street remains cynical, viewing the fund's near-collapse as a predictable lesson in the risks of excessive leverage.

• The New Stack explores a shift in evaluating non-deterministic AI coding agents, arguing that they should be graded on their actual output rather than as conversational chatbots. • The proposed methodology utilizes executable contracts and detailed scorecards to measure the functional success of the code produced by these agents. • This approach matters because traditional LLM benchmarks often fail to capture whether an agent can actually solve a complex software engineering task in a real-world environment.

Grid operator Neso can now respond with rolling blackouts as last resort without need for government powersBusiness live – latest updatesMillions of households could face power cuts at short notice under new rules to help the electricity system operator avoid widespread blackouts when energy supplies run low.The changes, which come into effect this week, will make it easier for the national electricity system operator to call for rotating blackouts at short notice if power supplies are running short. Continue reading...

Intervention follows days of delays and a near-miss between two planes on the Sydney aiport tarmac. Follow today’s news liveGood morning, and happy Monday. Here’s what’s on deck:Catherine King, the federal minister for transport, said she has asked Air Services Australia to come and talk to her today about a “pathway forward” after two aircraft were involved in a near-miss at Sydney airport on Sunday. Continue reading...