ZDNet AI
AI news source. Articles are auto-selected and adapted by Hamidun News editors.
Latest publications
OpenAI Costs: How to Set Spending Limits and Avoid Big Bills
OpenAI API costs can spike if an AI agent runs unchecked. Here's how to set Soft and Hard limits to avoid unexpected bills.
ExpressVPN updates ExpressKeys: password sharing and passkey support
ExpressVPN added secure credential sharing, passkey support, and improved device synchronization to its ExpressKeys password manager.
Fable 5 by Anthropic Sets Record on Freelance Tasks, but Cannot Replace Humans
Anthropic's Fable 5 model set a new record on the freelance work automation benchmark, surpassing competitors, but is not yet able to fully replace a human specialist.
AI tokens are driving up corporate cloud bills — ZDNet flags the trend
Large companies' spending on AI tokens is growing uncontrollably, echoing the chaos of early cloud pricing — only this time it's even harder to measure the value of the investment.
Startups become the main pathway to bigtech: how AI is reshaping hiring in IT
While major tech companies trim headcount under AI pressure, startups are opening hundreds of positions for engineers — and offering rapid entry into real AI development.
Chainguard launches Athena coalition: AI to patch open-source flaws before hackers do
Chainguard has assembled the Athena partner coalition, which uses AI to automatically find and fix vulnerabilities in open-source code—faster than attackers can find them.
Chrome, Edge and Firefox: AI Feature Test Reveals an Unexpected Winner
A ZDNET journalist tested Gemini in Chrome, Copilot in Edge and Firefox’s multibot — and chose Edge: Copilot analyzes all open tabs at once and ‘sees’ the screen via Vision.
Visa and OpenAI Created an AI-Agent Shopping System: How It Works and Whether to Trust It
Visa and OpenAI launched a partnership: ChatGPT agents can now independently make purchases — with virtual tokens, spending limits, and user approval.
AI Agents as Interns: Why Access Rights Need to Be Limited Upfront
Experts warn: AI agents act like enthusiastic interns — fast and without hesitation. Without clear access restrictions, this can result in serious and irreversible errors.
40% of companies ready to shut down AI-agent projects — three lessons from digital directors
Gartner forecasts: by 2027, 40% of corporate AI agents will be shut down. Three digital technology directors explained how not to end up among the failures.
Apple Limited AI Capabilities of Siri in iOS 27 for Older iPhone Models
In iOS 27, Apple will enhance Siri with AI capabilities, but support is limited to the latest iPhone, iPad, and Mac models due to computational power requirements.
How to Get New AI-Powered Siri — Join Apple's Waitlist
Apple has opened a waitlist for its new AI-powered Siri assistant. Learn how to join and when to expect early access.
Apple Announces Redesigned Siri and iOS 27 at WWDC 2026 Conference
At the annual WWDC 2026 developer conference, Apple unveiled an updated Siri with new capabilities and released the iOS 27 beta for developers.
Gemini in Android Auto: how the voice assistant changed my drives
After two months of using Gemini in Android Auto, I found how the voice assistant simplifies car control and makes daily drives more convenient and safer.
IBM and Red Hat invest $5 billion in automating open-source vulnerability management
Project Lightwell is an AI-based initiative to find and remediate vulnerabilities in open-source software at industrial scale, involving 20,000 engineers.
When Old Data Derails AI Deployment: Risks and Solutions
When deploying AI systems, companies rush to use all available data, including old archives. But they overlook the security and quality risks hidden within them.
Ollama Instead of ChatGPT: Saving Money, Privacy, and the Planet
A developer ditched ChatGPT for Ollama — a free local AI that works without internet, subscriptions, or data tracking.
Google Gives Users Control Over Sources in AI Overviews
Google added Preferred Sources feature to AI Overviews and AI Mode, allowing users to add favorite websites to AI search results.
Anthropic Released Opus 4.8 Focused on Model Honesty
Anthropic introduced Opus 4.8, a Claude model that better acknowledges its limitations and is ideal for complex coding tasks.
How to Be an Innovator in the Age of Agentic AI: The American Express Approach
An American Express executive has revealed 4 ways to stay ahead of competitors with autonomous AI agents.
Perplexity Launched Bumblebee — A Scanner for Finding Malware in Developers' Code
Perplexity launched Bumblebee, a tool for scanning code for known malware. This read-only solution is designed for development teams seeking to identify vulnerabilities following supply-chain attacks.
Anthropic Released Opus 4.8: Model Showed Equal Alignment with Claude Mythos
Anthropic introduced two new models — Opus 4.8 and Claude Mythos Preview — which demonstrated equal levels of safety and reliability in independent testing.
AI Is Leaving the Cloud: Dell on Hybrid Infrastructure of the Future
Dell recommends that companies move AI workloads to their own infrastructure due to rising cloud costs, data sovereignty requirements, and AI agents' need for minimal latency.
USA allocates $9 billion for Nvidia AI chips in race with China
The U.S. government plans to spend $9 billion on purchasing high-performance Nvidia chips to advance artificial intelligence technologies amid heightened global competition.
Google Gemini surpasses Siri in Android Auto
Google Gemini outperforms Siri in Android Auto: the voice assistant better recognizes driver commands, finds services faster, and integrates more seamlessly with in-car navigation.
Why I Abandoned Copilot and Chose Claude for Word, Excel, and PowerPoint
Claude AI now works directly in Microsoft 365 and helps create, edit, and analyze documents better than the built-in Copilot.
When an Ophthalmologist Makes a Mistake: How AI Helps Verify a Glasses Prescription
An ophthalmologist issued a correct prescription for distance vision but completely wrong for computer work. The author used AI to decode the numbers and discover the mistake.
Rust Will Protect Linux from AI-Generated Code Vulnerabilities, Says Greg Kroah-Hartman
The lead Linux kernel maintainer sees Rust as a tool to protect the operating system from vulnerabilities that AI systems may introduce when generating code.
96% of IT Professionals Have Adopted AI, but Validation of Results Remains the Key Challenge
New research shows that nearly all IT professionals are already using AI agents, but the most critical skill in this new cycle is validating AI results before production.
Google reveals Android Auto of tomorrow — and it makes old cars look ridiculous
At Google IO, Google showed an updated Android Auto with a new interface, maps, and Gemini integration — the demonstration was so impressive that the current state of your car seems hopelessly outdated.