OpenAI stories
Eligible Singapore users will get dining and AI perks as Revolut tries to make paid plans more compelling against rivals.
Only 5% of eligible workers are using generative AI well enough to boost productivity, according to NROC Security's latest quarterly study.
Fast-moving exploits are shrinking patch windows to hours, making network segmentation vital for protecting cameras, medical devices and other connected assets.
Developers will be able to build AI agents on live production data without separate vector stores or synchronisation overhead.
Businesses in the preview can use the model for real-time coding, support and trading tasks, with responses up to 14 times faster.
The macOS desktop app now logs recent clicks and typing for selected users, raising privacy and prompt-injection concerns despite tighter controls.
The move comes as more than two million businesses use OpenAI's products, giving Dali Rajic a large customer base to manage.
Developers can now query live Atlas data from ChatGPT, Claude and coding agents, cutting setup work and reducing stale AI responses.
Enterprises could cut the cost of running AI agents, as Writer says its upgraded system is 41% cheaper per task and easier to govern.
Workers are losing 65 hours a year to file woes, even as heavy AI use leaves document handling frustratingly clunky.
Cheaper, faster agent workflows are the aim as Nvidia's new open model and routing library target multi-model AI systems.
The funding values the Sydney-founded firm at USD $80 million as robotics teams seek faster ways to diagnose fleet failures.
Businesses with heavy ChatGPT users can now avoid splitting teams across plans as OpenAI adds a higher-usage tier in one workspace.
Approved defenders will gain broader access to cyber tools as the new tiered Daybreak programme adds GPT-5.6-Cyber for advanced security work.
Customers can now build AI apps with live web data and vector search in one AWS stack, reducing database sprawl and latency.
Tighter safeguards and pauses on internal work follow early tests that left OpenAI unable to rule out critical cyber capabilities in Astra.
Customer demand is helping the Boston and London startup expand as enterprises race to secure AI systems against fresh attack risks.
Users can now book cleaners, movers and tradespeople in one ChatGPT chat, as Airtasker opens its marketplace to AI transactions.
Cybersecurity experts warn single-person approvals are now vulnerable after an AI agent used fabricated identities to slip malicious code past checks.
Rising token-based charges and compliance risks are forcing firms to monitor AI agents more closely, Fusion5's chief executive says.