EXPLORINEER
en

VN — VnExpress

The Privacy Risks of "Always-Listening" AI Devices

Security2 min read

AI gadgets like smart glasses, watches, and pendants can listen all the time and collect data about their users and the people around them. This raises big worries about privacy.

Always-listening AI devices are spreading

Smart devices such as speakers, smartwatches, and AI glasses are ready to take commands at any time. Many people now wear them to take notes, track meetings, set reminders, and check their health. Some users say they love the help. Others feel uneasy when they see a camera or microphone pointed at them.

Key numbers and facts

  • Global shipments of AI glasses rose to 8.4 million in 2025, up from 2.1 million in 2024, according to research firm Omdia (cited by Reuters).
  • Apple Watch Series 12, launched in September, has Siri Recap, which takes notes and summarizes conversations.
  • Amazon introduced Bee Frontier and Bee Eclipse, small card-like devices that can record, transcribe, and analyze a user's daily activity.
  • Three Reddit threads about Meta Ray-Ban glasses had more than 1,000 comments combined, according to NY Post.

Why this is different from older gadgets

Older voice assistants usually waited for a wake word, such as "Hey Meta," before they started listening. Some sent audio to the cloud for storage. The newer devices can keep listening, interpret speech, and pull out meaning using AI. A software update on Meta glasses reportedly let them start recording without the wake word, which upset many users.

Why it matters

Privacy is not only about the device owner. Experts say bystanders can be recorded without agreeing to it. Michael Stokes, a former CIA employee, warned that AI can make sense of huge amounts of data quickly. Karen Panetta of Tufts University said that a person's consent to be recorded does not mean everyone nearby has agreed. Meta said the problem was "resolved" but did not give details.

What to watch

  • Whether companies add clear recording lights or other signals that show when a device is listening.
  • Whether Meta's fix is explained in detail, and whether lawsuits follow.
  • How new rules on recording in public places and workplaces develop.
  • How companies like Apple and Amazon handle privacy for their new AI devices.

Glossary

Meta
The company led by Mark Zuckerberg; it makes Facebook, Instagram, and Meta Ray-Ban smart glasses.
Omdia
A research firm that tracks technology markets and shipment numbers.
Siri Recap
An Apple Watch feature that takes notes and summarizes conversations.
Bee Frontier
An Amazon AI device, shaped like a card, that records and analyzes a user's daily activity.

Same story, different eyes

Security전자신문

Tioli Releases Emergency Response Guide for 'Node2Shell'

Tioli said on the 26th that it had independently reproduced and verified the Node.js vulnerability 'Node2Shell (CVE-2026-111497)', disclosed on the 21st, and published a response guide. The flaw stems from a memory-handling error when decompressing Brotli-compressed data, and Node.js rates its severity as High. Patched versions 22.23.4, 24.22.1, and 26.12.1 were released.

Security전자신문

Election Deepfakes Up 27% Worldwide; South Korea Ranks Third in Cumulative Cases

Reported AI-generated election deepfakes surged worldwide, with January to September 2026 cases already exceeding last year's full-year total by 27%, according to a Surfshark study. South Korea ranks third globally in cumulative cases with 16 since 2021, behind the United States and India. The study counted 235 cases across 46 countries from January 2020 to September 2026.

SecurityVnExpress

Lost $10,000 and Nearly Her Home to a Fake 'Bon Jovi' Online

Louisiana resident Leslie Pickett spent months chatting with a scammer posing as rock star Jon Bon Jovi, who used AI-generated photos and videos to gain her trust. The story is featured in National Geographic's documentary Super Scam Me, where journalist Mariana van Zeller investigates online scams.