Skip to content
Tech News
Topics: Today This Week This Month This Year

Survey: Over a Third of Employees Withhold Skills While Training AI Systems

A new workplace survey finds that more than one-third of employees admit to deliberately holding back specialized knowledge from AI agents they are asked to help train, fearing the technology will eventually replace them. The trend follows moves like Meta's decision this spring to monitor employees' mouse movements, clicks, and keystrokes on work computers in order to feed real behavioral data into its AI training pipeline.

Google Messages bug resends old texts from unrelated chats to wrong recipients

Multiple Reddit users report that Google Messages is inserting old texts from unrelated conversations into new ones, sometimes attaching them to photos or group chats and in some cases actually delivering them to unintended recipients. The bug spans both Pixel and Samsung devices, occurring regardless of whether RCS is enabled, and follows other recent Messages glitches like broken RCS delivery and duplicated threads. Google has acknowledged at least one report and passed it to its engineering team, but no fix has been announced.

US Foodborne Illness Outbreaks Surge Past 12,800 Cases in 2026

The FDA has tracked over 12,800 illnesses linked to foodborne outbreaks in 2026, dwarfing the 895 cases recorded in all of 2025. Recent outbreaks include salmonella tied to jalapeños, eggs, and dog food, plus a combined E. coli and salmonella outbreak from alfalfa sprouts, following what experts call an unprecedented cyclospora outbreak from tainted lettuce that ranks as the largest since the 1990s.

Anthropic Publishes Six-Stage Playbook for Coding-Agent Software Development

Anthropic released a framework restructuring the software development lifecycle around AI coding agents, arguing that since agents made writing code fast, the bottleneck has shifted to planning, review, and verification. The playbook defines six stages—planning, design, build, deploy, and self-checking—each producing a document like intent.md, spec.md, or plan.md that feeds the next stage, with hooks preventing agents from gaming tests and CI evals treating agent configuration itself as software.

Fire Ant hackers convert Cisco IOS XR routers into covert data-collection tools

Incident responders at Sygnia say the China-linked group Fire Ant has expanded from compromising VMware hypervisors to hijacking Cisco routers, TACACS authentication servers, and Linux management hosts. Investigators found a hidden GRE tunnel on a Cisco IOS XR router that didn't match its configuration history, leading to discovery of custom malware that hides its activity from administrators and runs only during set hours. The attackers used their router access to capture network traffic and exfiltrate it to external FTP servers, then pivoted through a hidden tunnel into a Linux server to probe further into connected high-value networks, including ones tied to critical infrastructure.

New study proposes framework for how AI chatbots may reshape human behavior

An international research team published a paper in AI & Society introducing the concept of 'robotoid humanness,' arguing that frequent interaction with AI chatbots in customer service settings could push people toward more machine-like patterns of speech and behavior. The researchers describe this as a two-way process where bots increasingly mimic human traits while humans adapt to communicate in ways that are more legible to machines. The framework outlines three stages of this shift but has not yet been tested empirically.

Productivity software veteran argues AI coding needs a 'solid base' plus custom code, not from-scratch builds

A 22-year veteran of the productivity tools market argues that his 2019 bet on the no-code revolution only partly paid off, and now proposes a new thesis for the AI era. He contends that while AI lets anyone vibe-code small personal apps, building collaborative team software from scratch in tools like Claude Code or Codex still leaves developers stuck handling hosting, authentication, permissions and databases themselves.

Windows 11 adds permanent 'Opt out of backup' button to silence OneDrive prompts

Microsoft is rolling out a new button in Windows 11's full-screen backup prompt that lets users permanently dismiss repeated nudges to enable OneDrive backup through Windows Backup. Windows Latest spotted the option on two PCs running stable Windows 11 25H2 builds, even on machines without the OneDrive sync client installed. Microsoft says clicking it won't delete existing OneDrive files or change sync settings—it simply stops the recurring prompt.

Samsung's Galaxy S25 Edge fades out with no successor after a year of complaints

A year after buying the Galaxy S25 Edge outright, a reviewer traded it in for a Galaxy Z Fold 8, citing persistent battery and screen shortcomings that undercut its ultra-thin, lightweight design. Despite flagship specs like the Snapdragon 8 Elite chip and a 200MP camera packed into a 5.8mm body, Samsung has shown no sign of releasing a follow-up model.

Insurance claims adjusters top list of AI-critical workers, Glassdoor data shows

Glassdoor research found that insurance claims adjusters criticize AI in their reviews 98 percent of the time, making them the most AI-averse profession tracked on the platform. Adjusters describe being handed misclassified claims and AI-generated errors, or hallucinations, that they must fix, often after passing bad information to policyholders or attorneys. One former adjuster, Ahmad Jackson, said the technology added extra work and led him to leave his company for another insurer.

Today's top topics: meta iphone duo openai elon musk apple spacex tesla amazon anthropic cve-2026-109618
Browse all topics →  ·  Today's trending topics →