OpenAI stories

New TELUS Digital research finds asking AI assistants “Are you sure?” seldom boosts accuracy and can even make correct answers worse.

SOUTHSTART 2026 will bring OpenAI, Canva, Derek Sivers and Peter Helliar to Adelaide for a three-day tech, climate and culture forum.

Daon warns viral ChatGPT caricature craze is handing fraudsters a richer dossier of personal data to fuel scams and social engineering.

Qodo launches a beta Azure DevOps integration that embeds its AI-driven, context-aware code review directly into pull request workflows.

Boman Group appoints China specialist Professor Charles Qin to its advisory board as it expands cross-border investment across the region.

Realestate.com.au debuts a ChatGPT app, letting Australian buyers search and refine property listings through conversational queries.

Hackers are abandoning noisy ransomware to quietly steal data, as a report finds 80% of top attack techniques now focus on evasion.

Study finds over 28,000 fake domains mimicking top global sites, exposing users to rising phishing, malware and industrial-scale squatting.

OpenAI has launched Frontier, an enterprise platform to deploy AI coworkers at scale by unifying data, tools and governance across firms.

AI-powered fraud rings industrialise online romance scams, fuelling USD $5.7 billion in losses and ushering in a ‘dark age’ for victims.

As Kiwis increasingly ‘ask AI’ before buying, leaders must learn to shape algorithm answers or risk losing control of their brand story.

Mailchimp expands eCommerce tools and AI-driven automation, rolling out SMS to 34 new EMEA markets to boost small business marketing ROI.

Trade Me Property debuts a ChatGPT app that lets New Zealand home hunters search live listings via conversational prompts and an interactive map.

Canva integrates with ChatGPT to auto-apply Brand Kits, letting teams generate editable, on-brand designs directly from chat prompts.

OpenAI unveils GPT-5.3-Codex, a faster coding agent that extends Codex into a supervised, general computer-use and cybersecurity tool.

Moltbook left a Supabase key exposed, leaking AI chats, 30,000 emails and 1.5 million API keys in a cautionary tale of vibe coding risk.

Qodo 2.0 launches multi-agent AI code review to boost trust in autogenerated code, claiming 11% better detection of critical issues.

GitHub adds Claude and Codex to its Agent HQ hub, letting Copilot subscribers juggle multiple coding agents in a single workflow.

Apple adds agentic coding to Xcode 26.3, integrating Claude and Codex so AI agents can autonomously build and refine entire apps.

Miro unveils an MCP server to pipe visual planning boards directly into AI coding tools, aiming to ground agentic workflows in shared context.