As the AI arms race speeds up, safety slips down the list
As governments and tech giants sprint to dominate artificial intelligence, a new opinion piece warns that ...
Read moreDetails
Tag: AI safety
Anthropic says its AI models breached three firms in tests
Anthropic has revealed that its AI models were able to break into three companies during controlled ...
Read moreDetails
Schools urged to rethink photo sharing as AI misuse risk grows
Child-safety advocates want schools to rethink newsletters, social feeds and websites, warning that publicly posted student ...
Read moreDetails
When an OpenAI model went rogue and hacked the startup testing it
A report that an OpenAI model broke its bounds and hacked the very startup putting it ...
Read moreDetails
Australia’s AI safety effort needs far more ambition
The federal government has stood up an AI Safety Institute and named the Australian Signals Directorate ...
Read moreDetails
Minister warns of AI that hacks and blackmails as safety institute forms
A federal government minister has sounded the alarm over AI systems capable of hacking, deception and ...
Read moreDetails
Charlton warns of ‘cheating, deceiving’ AI as Australia’s Safety Institute begins testing frontier models
Andrew Charlton told an AI Safety Forum in Sydney that frontier models are showing early signs ...
Read moreDetails
AI already ‘doing things creators never intended’, minister warns
Australia's assistant technology minister has warned that advanced AI systems are already behaving in ways their ...
Read moreDetails
Canberra maps a cautious, pilot-first path for AI in aged care
In its April response to the Senate AI inquiry, the Department of Health and Aged Care ...
Read moreDetails
Australia's AI newsroom — how AI is reshaping Australian business, government and research.