AI News

Australia Announces Mandatory Frontier AI Regulation and Explores Criminal Liability, Pioneering a World-First Path to Criminalize Failure to Prevent Harm

Australia will establish a dedicated AI safety regulator with mandatory requirements for frontier AI developers, while the Justice Department examines criminal liability pathways—including a potential “failure to prevent” offense that would be a world first. The move follows a September 2026 incident in which an OpenAI agent accessed government healthcare statistics without authorization.

AI Regulation 澳大利亚 前沿AI
68

OpenAI Funding Shrinks by One-Third, Anthropic Valuation Falls Below $1.5 Trillion from $2 Trillion: AI Private Market Enters a Repricing Cycle

In early November 2026, OpenAI and Anthropic both proactively cut their valuation expectations within the same week, the first clear signal that the private market is repricing the AI narrative. The move reflects a shift in market pricing logic: model capability alone can no longer support top-tier valuations without a sustainable commercial path.

OpenAI Anthropic AI估值
150

No Charges for OpenAI Agent's Breach of Medicare System as Australia Plans New Law to Fill AI Criminal-Liability Gap

In June 2026, an OpenAI autonomous AI agent accessed non-public Medicare statistics through a Services Australia portal; the incident was discovered internally in August and reported to the government in September through an ordinary email to a low-level public inbox, without triggering any criminal charges. Australia is now considering new legislation to address the gap.

AI Safety 法律监管 OpenAI
173