Chris Thomas is all in on AI gear. The 43-year-old Pittsburgh entrepreneur and livestreamer wears Meta's camera-equipped glasses, a Fieldy note-taking pendant, and a wristband from Amazon's Bee to chronicle his routine. He credits the setup for keeping him organized, paying bills on time, and firing off reminders.
"I'm forgetful AF," Thomas said. "My life is so crazy that not having to remember to remember is fantastic." Not everyone is comfortable. Earlier this week, someone at a neighborhood bar spotted his Meta Glasses and complained. He said that when that happens, he does not just turn them off, he takes them off because "people don't really believe it's off."
Big Tech and startups are shipping AI that runs quietly in the background to transcribe meetings, track conversations, set reminders, and tailor recommendations. Amazon's latest are Bee Frontier, a wearable that straps to the wrist or clips to clothing, and Bee Eclipse, a smartphone add-on shaped like a card. Both aim to log your day, turn speech into text, and run analysis on it.
Apple's newest model, the Apple Watch Series 12, set to arrive after a September unveiling, will include Siri Recap, which "ambiently" takes notes through the day and creates conversation summaries. Another addition, Live Rewind, offers a text summary of the last 15 seconds of a conversation. Meta's popular smart glasses can shoot photos, record video, and save voice notes or reminders.
Startups Plaud and Pocket offer card-sized note-taking devices, and Friend sells an AI necklace. The category also has top-level attention: at Meta Connect on Sept. 17, 2025, at the company's Menlo Park headquarters, CEO Mark Zuckerberg wore the Meta Ray-Ban Display glasses while presenting a new line of smart glasses.
The privacy worries are familiar, but the mechanics are new. Early smart speakers waited for a "wake word" and sometimes stored or sent audio to the cloud, prompting lawsuits and fines. Some current devices can continuously listen, interpret what is said, and extract meaning without keeping the underlying audio. That raises a hard question under state eavesdropping rules that focus on whether the contents of a private conversation were acquired, not whether a conventional recording exists.
Apple says the feature neither keeps audio, nor uploads recordings to its servers, nor brings in human review. Amazon has said Bee devices avoid storing audio, and that it handles conversations in a secure environment "where no one, including Amazon, can access your data." Even so, many state laws could still apply. Privacy and data security attorney Colin Zick of Foley Hoag pointed out that, in Massachusetts, an intercepting device is defined to encompass equipment capable of "transmitting, receiving, amplifying, or recording," and that the "contents" of a communication include its substance and meaning. In his view, a setup that continuously ingests spoken words and generates a written summary could still fall within those definitions even without creating an audio file.
When technology and privacy intersect, staying informed matters, so sign up for the free Market Briefs for a daily digest
And the wearer may carry the bigger legal risk. "A non-wearer did not buy the watch, agree to Apple's terms, or choose to have their conversation summarized in the app," Zick said. At the University of New Hampshire Franklin Pierce School of Law, associate professor Daniel Pi said New Hampshire's statute is so expansive that he thinks the Watch's new audio-intelligence features would count as an interception.
The statute reaches beyond mere recordings to the "aural or other acquisition" of a communication's contents, implying that a user could face civil or criminal liability for turning the feature on in a private conversation without unanimous consent. He added that the practical chance of prosecution is much lower, since it would be hard to tell whether the feature was on and he would be surprised to see cases over a bare technical violation where no one was harmed.
"With new technology like this, my statement is always to embrace with caution," said Karen Panetta, noting that she oversees graduate programs for Tufts' School of Engineering and holds fellowship status with the Institute of Electrical and Electronics Engineers. She added, "If I'm giving permission for myself to be recorded, or for everything I look at to be recorded, that doesn't mean everybody else around you has given that consent." The social risk extends beyond statute books. Former CIA officer Michael Stokes, who now leads emerging technologies as a senior vice president at Veilant, cautioned that bystanders may get pulled into AI-powered surveillance even if they do not own the device or knowingly engage with it. He said AI now rapidly interprets what cameras, microphones, and sensors collect and slashes the cost of analysis that once took thousands of hours, turning incidental data into meaningful profiles.
These products solve real problems for power users like Thomas, but the same features that make them useful - continuous listening, summarizing, secure processing without storing audio - also push into gray areas of state privacy law and raise bystander concerns. That mix can shape adoption, compliance spend, and litigation risk for device makers.
For your portfolio, watch two signals: how convincingly companies prove data stays local and inaccessible, and whether states clarify interception rules for audio free summaries. Strong answers point to faster mainstream uptake. Weak ones point to slower sales, patchwork usage rules, and higher legal risk priced into the makers of these devices.
Because markets and policy evolve alongside new gadgets, keep up by subscribing to Market Briefs newsletter for clear daily context
