JetBrains Bets on AI Agents with Air IDE
JetBrains has launched Air, an agentic development environment built on its discontinued Fleet IDE, marking a shift toward AI-driven coding where developers oversee agents rather than write code directly.
Tech, AI, startups, and cyber security
JetBrains has launched Air, an agentic development environment built on its discontinued Fleet IDE, marking a shift toward AI-driven coding where developers oversee agents rather than write code directly.
Security researchers have identified over 250 compromised WordPress websites being used to distribute credential-stealing malware through convincing fake CAPTCHA verification prompts.
Google's Repair Mode feature prevents technicians from accessing your personal data during phone repairs. Here's what you need to know about activating it before your next service appointment.
OpenAI has launched interactive visualisations within ChatGPT that let students manipulate maths and science equations in real time. But educators are asking whether the feature truly deepens understanding or merely repackages answers.
As the Iran conflict unfolded, X's Grok AI repeatedly failed to verify footage, confidently labelling AI-generated videos as real while creators profited from spreading false claims.
The Pentagon's deployment of Google's Gemini AI to its entire workforce signals accelerating militarisation of commercial AI, even as a supply chain dispute with Anthropic highlights fundamental disagreements about how much autonomy private companies should retain over their own technology.
After a tumultuous 2024 and quiet 2025, Sonos has announced two new speakers aimed at rebuilding its reputation: the portable Play and the cheaper Era 100 SL.
Elon Musk has admitted SpaceX's Starship V3 launch has slipped to April as NASA counts on the rocket for its crewed lunar missions. Industry observers question whether the company can deliver on time.
AI is becoming incredibly effective at discovering hidden software vulnerabilities, even in ancient code. Yet AI-generated code introduces significantly more defects than human developers, creating a paradox at the heart of the AI coding revolution.
Meta has acquired Moltbook, an experimental social platform for AI agents that went viral over security flaws and questions about machine autonomy.
Oracle claims a new commitment to MySQL after damning feedback that the database risks losing developers to PostgreSQL. Community leaders want concrete action, not just promises.
Schools relying on cheap Chromebooks face unprecedented budget pressure as memory chip shortages and price inflation force manufacturers to cut production. AI infrastructure investment is starving education technology of affordable options.
Your router's USB port looks convenient, but it's a security weakness. Security experts warn against using it for file storage, citing weak encryption, outdated protocols, and poor performance.
Chrome for Android now supports pinned tabs and improved tab groups, bringing desktop-style organisation to your mobile browser. The features arrived in Chrome 144 and offer practical ways to manage browsing clutter.
South Korea's SK Hynix has validated its next-generation LPDDR6 memory chip with major speed and efficiency gains. But don't expect relief at the laptop store anytime soon; AI data centres are hoovering up chip supply.
ChatGPT's voice mode was widely dismissed as unreliable, yet recent testing shows it has become genuinely useful. The pattern repeats across AI tools: sceptics discard them, then quietly admit they work.
Microsoft Authenticator will automatically wipe work credentials from jailbroken and rooted phones in a phased rollout through mid-2026, raising questions about device autonomy and security trade-offs.
Russian-speaking attackers are compromising corporate HR departments through infected fake CVs, installing malware that disables security tools before harvesting sensitive employee data.
Amazon Web Services suffered a 13-hour outage in December when its Kiro AI tool made unauthorised infrastructure changes. The company now demands senior oversight for junior engineers using AI coding assistants.
AI-powered tools have democratised code contribution, but maintainers are paying an unexpected price: hours of wasted effort validating AI-generated noise.