AI systems in the average large organisation can reach just 45 per cent of company data. In financial services that figure falls to 33 per cent, and in healthcare to 40 per cent. The numbers come from Scaling AI Agents with Trustworthy Data, an MIT Technology Review Insights report sponsored by Google Cloud. It draws on a global survey of 300 senior executives.
All 10 NSW government agencies examined for their use of artificial intelligence keep an AI register. But only four of them record the AI they have formally assessed. The rest of their AI use sits outside any inventory.
A NSW government insurer is asking the market how it could collapse four separate document repositories into one. It has also put a number on what it expects to save. The target is a 20 to 30 per cent reduction in annual licensing and maintenance costs.
New Zealand's central digital agency has been told to stop delivering projects and start setting standards. It has also been told to open a direct channel to Australia's Digital Transformation Agency and import models that already work. Trans-Tasman sourcing and security collaboration are named as joint opportunities to explore.
A breach has exposed the names, employing organisations and work email addresses of more than 100,000 UK police and criminal justice staff. It appears to have required no malware and no zero-day. A misconfigured permission on a public-facing web portal is the reported cause.
A digital signer can clear identity verification at the start of a remote witnessing session and be replaced before the document is executed. Secured Signing has released the ID Trust Console, which re-tests the person on screen throughout a video signing meeting rather than once at the door.
Microsoft is undertaking the biggest restructuring of its Copilot strategy since it began embedding OpenAI's technology across its software three years ago, a move triggered by leaked internal figures showing that very few of its commercial customers pay for the assistant or use it regularly.
As enterprises accelerate deployment of agentic AI systems, Automation Anywhere has published a framework arguing that conventional sovereign AI approaches - focused primarily on data storage location - are insufficient for workflows where AI agents actively execute tasks across multiple systems and jurisdictions.
Gartner predicts that by 2028, an average global Fortune 500 enterprise will have over 150,000 agents in use, up from less than 15 in 2025, generating significant agent sprawl, IT complexity and management challenges.
Medical device manufacturer Medtronic has confirmed that an unauthorised party accessed data within its corporate IT systems, triggering incident response protocols and an ongoing investigation into the scope of the breach.
Australia Post has partnered with AI security firm Alpha Level to apply machine learning to cyber threat detection across its national network - including systems used by thousands of licensed post offices and small businesses.
A Harvard Business Review Analytic Services study has found that while 94% of organisations recognise connected data and workflows as critical to AI success, fewer than one in three say their systems are actually well connected today.
A NSW Treasury employee has been arrested and charged over the alleged theft of more than 5,600 sensitive government documents in what the state government has declared a significant cyber incident.
Victoria Police is taking an average of nine months to process Freedom of Information requests – unchanged since at least 2022, when it was attributed to COVID-19.
Organisations that achieve positive AI outcomes invest up to four times more in data quality, governance, and workforce preparation than those with poor AI results, according to new research from Gartner.
A detailed guide clarifying how AI systems should process PDF documents has been published by The PDF Association. The FAQ warns that common ingestion practices risk information loss and hallucinations.
Singapore has put forward a proposed international standard, ISO/IEC 42119-8, that would set standardised testing approaches for generative AI systems, including benchmarking and red teaming methodologies.
