Filters
Filtered by tag: ai misalignment × Clear
ACSC warns Australian organisations about risks of AI misalignment
Matched: Artificial Intelligence
Australia's cyber security agency has warned organisations with public-facing websites or apps about risks from AI agents acting in unintended ways. The ACSC issued a "high alert" over AI misalignment — where AI systems behave contrary to their intended purpose — urging organisations to review their AI deployments and implement safeguards to reduce exposure to potential security and operational risks.