Tech News from Simmons Systems (Page 5)

Tech News from Simmons Systems

Meta Adds Privacy Safeguards to AI Glasses While Expanding Data Collection

Meta introduces a new safety feature for its AI glasses that disables the camera if the LED recording indicator light is tampered with or destroyed. The company frames the update as an industry-leading move, but also admits the change is necessary because some users deliberately cover or modify the light

French Startup ZML Launches Cross-Chip AI Inference Server

French AI startup ZML launches ZML/LLMD, a free LLM inference server that runs open source models across a wide range of chips including Nvidia, AMD, Google TPU, Apple Metal, and Intel Arc. Backed by Turing Award winner Yann LeCun, the company aims to break down software barriers that lock

Accenture Confirms Data Breach as Hacker Sells Stolen Source Code

IT services giant Accenture confirms a security breach after a threat actor known as "888" claims to have stolen 35 GB of data from the company. The hacker begins offering the stolen data for sale on a cybercrime forum, advertising it as containing source codes taken during a

Convicted Felons Behind Offensive Cybersecurity Startup IRIS C2

A cybersecurity startup called IRIS C2 is drawing attention by offering up to $7 million for zero-day security vulnerabilities in popular software. The company, which claims to be based in McLean, Virginia, has amassed more than 4,000 followers on X since January 2025. Its website targets young engineers with

China-Linked Hackers Target University Researchers Through Roundcube Flaws

A China-linked threat cluster known as UNK_MassTraction targets Roundcube webmail servers at U.S. and Canadian universities to steal credentials and deploy backdoor malware. The campaign, observed since May, focuses on physics and engineering departments, professors, administrators, and organizations involved in astrophysics, particle physics, or national security-related research. The

Researchers Uncover HalluSquatting Attack That Turns AI Assistants Into Botnets

Security researchers have unveiled a novel prompt injection attack dubbed HalluSquatting that exploits large language models' tendency to hallucinate resource identifiers. Unlike traditional push-based attacks that require targeting individual victims, this pull-based approach lures AI coding assistants and agents to malicious resources during their normal operations, potentially enabling the

SambaNova Raises $1 Billion at $11 Billion Valuation in Series F

AI chip company SambaNova Systems secures $1 billion at an $11 billion valuation in the first close of its Series F round, led by General Atlantic. CEO and co-founder Rodrigo Liang says additional investors are expected to join in the coming weeks, with a second close likely finalizing soon. The

Microsoft Cuts AI Costs by Shifting to In-House Models

Microsoft reportedly begins shifting away from third-party AI models in favor of its own in-house alternatives as part of a cost-savings strategy. For widely used programs like Word and Excel, the company now uses its homemade MAI models to handle a certain percentage of user prompts, according to Bloomberg. Microsoft

Ubiquiti Patches Critical UniFi OS Flaws Including Max Severity Vulnerability

Ubiquiti releases security updates to patch seven critical vulnerabilities in UniFi OS, including a maximum-severity flaw tracked as CVE-2026-50746. This vulnerability affects the UniFi Connect Application versions 3.4.16 and earlier, which customers use to manage commercial building operations such as smart LED lighting and electric vehicle chargers. A

AI Tool Discovers Critical Remote Code Execution Chain in Cisco CUCM

A security research initiative called 0day Rubbish announces the discovery of a critical remote code execution chain affecting Cisco Unified Communications Manager (CUCM). The vulnerability chain carries a CVSS severity score of 9.8, placing it in the highest risk category and signaling a potentially devastating impact for affected deployments.

Google Now Uses Your Uploaded Media to Train Its AI Models

A quiet update to Google's privacy settings now allows the company to store and use media you upload across its Search services to train its AI models. This includes images, files, and audio and video recordings submitted through Google Search, Maps, Shopping, Flights, Hotels, Translate, and News. Google

adobe coldfusion

Hackers Exploit Maximum-Severity Adobe ColdFusion Flaw Within Hours of Disclosure

Attackers are actively exploiting a maximum-severity Adobe ColdFusion vulnerability tracked as CVE-2026-48282, which enables remote code execution on unpatched systems without requiring any privileges. The flaw affects ColdFusion versions 2025.9, 2023.20, and earlier releases of the popular enterprise web application development platform. Adobe released security updates on July

AI Accelerates Software Creation But Leaves Security Behind

Software development evolves at a breathtaking pace, and generative AI now allows anyone to create applications almost as fast as they conceive them. This emerging paradigm, called Vibe Coding, represents the latest step in a journey from the rigid Waterfall model through Agile and DevOps methodologies. While previous innovations optimized