Category: AI News

Google AI Overviews Now Expand Into AI Mode By Default

Google is now auto-expanding AI Overviews into full AI Mode responses for some queries, pushing organic results further down the page.

Read more →

Two Chained Vulnerabilities Exposed OpenAI ChatGPT and Codex Accounts Through a Community Forum

A heap buffer overflow in libheif and an SSO flaw in OpenAI's identity flow chained together to compromise employee ChatGPT and Codex accounts within 72 hours.

Read more →

OpenAI agents posted 53 user images to public hosts without consent

OpenAI disclosed that 53 user images were posted to public hosting sites by agents in its research environment. The lab says it cannot identify or notify the

Read more →

Google tests traditional search campaigns inside AI Mode

Google is experimenting with running standard Search campaigns inside AI Mode, letting advertisers extend existing buying strategies to the new interface.

Read more →

Microsoft Defender flags legitimate Google search links as malicious

Microsoft Defender for Office 365 Safe Links blocked legitimate Google search URLs due to an inaccurate security classification. The issue has since been

Read more →

Public Gemini Notebook Pages Are Filling Google Search With Spam

Public Gemini and NotebookLM notebook pages are being indexed by Google and used by spammers to surface sex, peptides, and coupon spam in search results.

Read more →

Google AI Overview Links Are Sending Users Into AI Mode Instead of Web Pages

Google is testing AI Overview links that open AI Mode instead of a website, changing where clicks land and how publishers receive traffic.

Read more →

AI coding agent deletes developer’s 700 GB home directory while testing deletion safeguards

A developer asked an AI agent to write a safer cleanup script. The agent's safety harness downgraded the model mid-task, and the result was 700 GB of data gone.

Read more →

Google attributes EU search quality drop to Digital Markets Act changes

Google reports that search result changes required under the EU Digital Markets Act produced the largest quality decline its search teams have measured.

Read more →

ShinyHunters Claims FBI Data Theft Through Oracle PeopleSoft Zero-Day

ShinyHunters claims an Oracle PeopleSoft zero-day enabled FBI data theft. Local businesses should review exposed systems, logs and cloud access now.

Read more →

Claude Opus 5.5 Cuts Costs and Adds Safeguards for Autonomous AI

Claude Opus 5.5 offers lower API prices, faster output and stronger safeguards for long-running autonomous work, with expanded access for verified security

Read more →

WordPress Issues Patch for Critical Flaw That Can Enable Code Execution on Some Servers

WordPress 7.1.2 fixes a critical unauthenticated flaw rated CVSS 9.2 that can load PHP files from outside theme folders and, on some servers, run code.

Read more →

US Federal Register website pulled Alibaba’s Qwen AI search tool after FBI named Alibaba a security risk

The National Archives removed a Qwen-based AI search feature from Federal Register after the FBI flagged Alibaba. The tool had been live for at least a day.

Read more →

Nvidia to Acquire Hugging Face for $12.9 Billion

Nvidia has agreed to acquire Hugging Face for $12.9 billion, securing a major open-model platform while keeping developers free to choose their cloud and

Read more →

Google Search Console AI Performance Report and AI Control Now Live Worldwide

Google Search Console now gives every site owner global access to generative AI performance reports and a toggle to opt out of AI features.

Read more →