Product1 distinct publisher3 min readUpdated
A Target worker said the stop command out loud and got feigned confusion. The clip ran anyway, to roughly 555,000 views across two platforms.
The Product Desk · Product desk

Compiled by The Product DeskSomething wrong?How this is made
In August, Toru Hinkle was stocking shelves at Target when two customers asked for a price check, then kept asking after being told the item cost $20 [1]. Hinkle spotted the blinking light on the men's Ray-Ban Meta glasses and worked out that the repetition was the content [2].
What happened next is the part every retail and service operator should read as a process failure rather than a bad-customer story. Hinkle called over a store manager, and the pair antagonised her too, asking the price again and deliberately misgendering Hinkle as "ma'am" and the female manager as "sir" [3]. Hinkle then tried the only lever the device appears to offer a bystander, saying "Meta, stop recording" out loud in the hope of triggering a voice control setting [4]. The customers pretended to be confused by the accusation that they were filming [5]. They were filming, and the exchange became social media content [6]. One clip viewed by The Verge had almost 400,000 views on Instagram and 155,000 on TikTok [7][8], roughly 555,000 across the two [9]. Another Target manager found the video while scrolling, without looking for it [10].
The store-level response worked, slowly and by accident: another customer called the police and the two men were escorted out [11]. Hinkle told The Verge the confrontation produced a pit in their stomach and that they stayed quiet for the rest of the shift [12]. Target declined to comment [13]. In one of the posted videos, the glasses wearer is seen dancing and fist-bumping a police officer who arrived at the store, with the officer saying "You boys stay out of trouble" [14].
Platform recourse was also reactive. The Verge reports the video is the exact kind of content Instagram has said it will now remove, and that the uploading account was full of similar prank videos harassing workers [15][16]. Instagram banned the account, and removed the Hinkle video with it, after The Verge asked about it [17]. That is enforcement triggered by a reporter, not by the employee, the manager, or the employer.
The structural problem is that ordinarily a person being recorded can leave the area or push back, and a worker on shift cannot [18]. Many people are meeting smart glasses for the first time at work, in public-facing jobs, and many workplaces have no policy for the device yet [19][20]. Resistance costs more when the job pays the rent [21]. Emily, who works at a Seattle tourist destination, told The Verge that being background in a video used to be part of the job, and that the worst case now is a much taller man standing over her filming with Meta glasses; she says a phone held at that angle could be challenged, while the glasses make it much harder to tell what is being filmed [22][23][24].
Meta has sold millions of pairs of its AI-powered glasses over several years [25]. Watch whether employers publish an actual script for this: who a worker escalates to, what a manager is authorised to say, and whether refusing service to a recording customer is protected. Absent that, the default policy is the one Hinkle used, which did not work.
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
At one point Hinkle attempted to stop the recording by saying "Meta, stop recording" out loud, hoping to trigger a voice control setting.
The customers pretended to be confused by the accusation that they were recording Hinkle.
The Verge reports the video made of Hinkle and their manager without consent is the exact kind of content Instagram has claimed it will now remove.
The account that uploaded the video was filled with similar prank videos harassing workers at their jobs.
Meta has sold millions of pairs of its AI-powered smart glasses over the last several years, and backlash to the devices has grown alongside sales.
In August, Toru Hinkle was stocking shelves at their job at Target when two customers requested a price check on an item and kept asking for the price again and again even after Hinkle told them the item was $20.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Well-sourced anecdotes, unquantified generalisations
The specific incident is strong for a single-outlet report: a named worker on the record, direct quotes, view counts the reporter verified, and a platform action The Verge triggered and confirmed. Against that, the employer declined to comment, Meta is not quoted on whether any bystander stop command exists, no legal or policy expert appears, and the wider claims about missing workplace policies and structural power imbalance are reporter framing with no survey or employer sample. One publisher, two worker accounts, no independent corroboration.
Devices are in the field, exposure data is anecdotal
Adoption of the hardware itself is real but only loosely evidenced here: an unattributed 'millions of pairs' sales characterisation, plus one worker's estimate of up to ten glasses wearers an hour at a high-traffic tourist site. Downstream adoption of the harmful behaviour is shown by one account full of worker-prank videos and a clip reaching roughly 555,000 views. There is no market data, no employer policy adoption, and no platform enforcement volume, so the pattern is demonstrated but not sized.
Headline framing runs slightly ahead of the sourcing
The reported incidents are concrete and the platform-enforcement finding is genuinely new, so this is not a hype piece. The gap is modest and upward: 'workplace menace' and the claim that the stop command 'is not a control' generalise from one failed spoken command and two worker interviews, without Meta confirming what voice controls exist, without employer or legal input, and without any measure of how common such encounters are.
Engagement and reputational incentives visible on all sides
The material itself exposes incentives: prank accounts monetise worker humiliation through reach, with one clip at roughly 555,000 views and an account stocked with similar videos; Instagram's stated removal policy was enforced only after a journalist asked, a reputational trigger rather than a systematic one; Meta's tamper mitigation is described as in progress amid growing backlash to a product line it is selling in volume. The publisher is a commercial tech outlet whose framing choices also serve engagement, and Target's refusal to comment is itself an incentive-shaped silence.
Confident on the incident, cautious on the trend
Confidence is high that the described encounters, the failed spoken stop command, the view counts, and the post-inquiry account ban happened as reported, since they come from a named worker, quoted interviews, and a reporter-verified platform action. Confidence is materially lower on the generalised claims about employer policy vacuums, structural power imbalance, and 'millions' of units, all of which are single-outlet framing without external data, employer comment, or legal grounding.
product
Meta blocked the search terms and left the videos up: nine takedowns from a list of 20-plus2 distinct publishers
product
Cinemas, classrooms and ICE: smart glasses now need a venue-policy contingency1 distinct publisher
leadership
Meta's glasses crackdown removed nine videos after a reporter sent the links1 distinct publisher
invest
Meta's Pay Structure, Not Its Policy Page, Is What the States Put on the Stand1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 20, 2026