AI Safety Stories - Page 19

Australian Stories

AI Safety stories

UST partners Anthropic to train 20,000 staff on Claude

The deal will embed Claude across UST's client systems and internal workflows, as the services firm trains 20,000 staff worldwide on the AI model.

By Sean Mitchell • 4 min read • Last month

Digital Transformation

Kinetic IT launches AI engineering division with Sete

The move targets government and critical infrastructure clients seeking secure AI deployment inside complex operations rather than advisory support.

By Mark Tarre • 3 min read • Last month

Global Stories

AI Safety stories

Robotics

AI agents emerge as top cyber threat, Exabeam finds

Security teams now see autonomous AI as a bigger internal danger, even as most say it is boosting productivity, a survey found.

By Mark Tarre • 3 min read • Last month

Data Protection

John Margerison on the new class of employee: AI managers

Businesses should treat AI like a new hire, as weak oversight could expose sensitive data and leave staff needing fresh skills to stay relevant.

By Jake MacAndrew • 3 min read • Last month

DevOps

Google sets out 20-question framework for AI agents

The framework aims to help IT leaders control security, governance and costs as agent-based systems move from pilot projects into production.

By Sean Mitchell • 5 min read • Last month

Software-as-a-Service

Molted touts operating layer for AI agents in production

As fleets of autonomous agents move into daily use, teams are wrestling with crashes, credentials and recovery that model demos never cover.

By Joseph Gabriel Lagonsin • 5 min read • Last month

Digital Transformation

Applause names Salvi Chief Executive as AI push grows

The software tester is expanding into agentic AI validation as Aatish Salvi takes over from Chris Malone and Tacita Morway becomes Chief Technology Officer.

By Joseph Gabriel Lagonsin • 4 min read • Last month

Data Protection

Adding AI to a platform is easy, adding it without breaking your security isn't

Most firms rushing AI into sensitive systems lack basic access controls, leaving customer data exposed to wider breaches and governance gaps.

By Justin Smith • 5 min read • Last month

Platform-as-a-Service

AWS adds Claude Sonnet 5 as it trims older services

New cloud features and lifecycle cuts could reshape AWS customers' AI and operations plans as older services head towards maintenance or sunset.

By Mark Tarre • 5 min read • Last month

Phishing

Google unveils Gemini 3.5 Live Translate & Android 17

Android users will get new security and multitasking tools first, as Google rolls out Gemini 3.5 Live Translate and other updates.

By Mark Tarre • 5 min read • Last month

Data Protection

Cloud Security Alliance adds AIUC-1 to STAR Registry

Buyers will gain a clearer signal on autonomous AI, as certified providers can now display a trustmark in the widely used STAR Registry.

By Joseph Gabriel Lagonsin • 4 min read • Last month

Data Protection

Anthropic AI controversy raises cyber trust concerns

Hidden tracking markers and a US standoff over a vulnerability-finding model are fuelling fears that AI now carries cyber and national security risks.

By Mark Tarre • 4 min read • Last month

DevOps

HP expands OpenAI partnership to scale AI across business

The rollout could speed up support, security fixes and software delivery across a business that channels more than 80% of sales through partners.

By Sean Mitchell • 4 min read • Last month

Malware

Anthropic widens Fable 5 safeguards & rates jailbreaks

New safeguards will let Fable 5 block more harmful cyber prompts, as Anthropic also seeks a common scale for jailbreak risk.

By Sean Mitchell • 4 min read • Last month

Data Protection

Most firms hit by AI security incidents, study finds

Most organisations are exposed to AI security breaches, with AvePoint finding 88.4% suffered at least one incident in the past year.

By Sean Mitchell • 4 min read • Last month

Stories from Other Regions

AI Safety stories

Digital Transformation

Canadian AI consortium launches: Telus, Scotiabank partners

Regulated firms in Canada can now share AI controls and intellectual property, with the first system already handling more than two trillion tokens a month.

By Jake MacAndrew • 3 min read • Last month