Two former Harvard students are launching a pair of “always-on” AI-powered smart glasses that take heed to, record, and transcribe every conversation after which display relevant information to the wearer in real time.
“Our goal is to make glasses that make you super intelligent the moment you place them on,” said AnhPhu Nguyen, co-founder of Halo, a startup that’s developing the technology.
Or, as his co-founder Caine Ardayfio put it, the glasses “offer you infinite memory.”
“The AI listens to each conversation you’ve and uses that knowledge to inform you what to say … kinda like IRL Cluely,” Ardayfio told TechCrunch, referring to the startup that claims to assist users “cheat” on all the pieces from job interviews to high school exams.
“If anyone says a posh word or asks you a matter, like, ‘What’s 37 to the third power?’ or something like that, then it’ll pop up on the glasses,” Ardayfio added.
Ardayfio and Nguyen have raised $1 million to develop the glasses, led by Pillar VC, with support from Soma Capital, Village Global, and Morningside Enterprise. The glasses can be priced at $249 and can be available for preorder starting Wednesday. Ardayfio called the glasses “the primary real step towards vibe considering.”
The 2 Ivy League dropouts, who’ve since moved into their very own version of the Hacker Hostel within the San Francisco Bay Area, recently caused a stir after developing a facial-recognition app for Meta’s smart Ray-Ban glasses to prove that the tech could possibly be used to dox people. As a possible early competitor to Meta’s smart glasses, Ardayfio said Meta, given its history of security and privacy scandals, needed to rein in its product in ways in which Halo can ultimately capitalize on.
Techcrunch event
San Francisco
|
October 27-29, 2025
“Meta doesn’t have a terrific fame for caring about user privacy, and for them to release something that’s all the time there with you — which obviously brings a ton of utility — is just an enormous reputational risk for them that they probably won’t take before a startup does it at scale first,” Nguyen added.
And while Nguyen has some extent, users may not yet have a very good reason to trust the technology of a few college-aged students purporting to send people out into the world with covert recording equipment.
While Meta’s glasses have an indicator light when their cameras and microphones are watching and listening as a mechanism to warn others that they’re being recorded, Ardayfio said that the Halo glasses, dubbed Halo X, don’t have an external indicator to warn people of their customers’ recording.
“For the hardware we’re making, we would like it to be discreet, like normal glasses,” said Ardayfio, who added that the glasses record every word, transcribe it, after which delete the audio file.
Privacy advocates are warning concerning the normalization of covert recording devices in public.
“Small and discreet recording devices will not be latest,” Eva Galperin, the director of cybersecurity on the Electronic Frontier Foundation, told TechCrunch.
“In some ways, this seems like a variation on the microphone spy pen,” said Galperin. “But I feel that normalizing using an always-on recording device, which in lots of circumstances would require the user to get the consent of everyone inside recording distance, eats away on the expectation of privacy we’ve got for our conversations in every kind of spaces.”
There are several states within the U.S. that make it illegal to covertly record conversations without the opposite individuals’ consent. Ardayfio said they’re aware of this but that it’s as much as their customer to acquire consent before using the glasses.
“We trust our users to get consent in the event that they are in a two-party consent state,” said Ardayfio, referring to the laws of a dozen U.S. states that require the consent of all recorded parties.
“I might even be very concerned about where the recorded data is being kept, the way it is being stored, and who has access to it,” Galperin added.
Ardayfio said Halo relies on Soniox for audio transcription, which claims to never store recordings. Nguyen claimed when the finished product is released to customers, it would be end-to-end encrypted but provided no evidence of how this could work. He also noted that Halo is aiming to get SOC 2 compliance, which implies it has been independently audited and demonstrates adequate protection of customer data. A date for the finished SOC 2 compliance was not provided.
Still, the 2 students will not be latest to privacy-invasive controversial projects.
While still at Harvard last yr, Ardayfio and Nguyen developed I-XRAY, a demo project that added facial-recognition capabilities to the Meta Ray-Ban smart glasses, demonstrating how easily the tech could possibly be bolted onto a tool not meant to discover people.
The duo never released the code behind I-XRAY, but they did test the glasses on random passersby without consent. In a demo video, Ardayfio showed the glasses detecting faces and pulling up personal information of strangers inside seconds. The video featured reactions of people that were doxed.
In an interview with 404 Media, they acknowledged the risks: “Some dude could just find some girl’s home address on the train and just follow them home,” Nguyen told the tech news website.
For now, Halo X glasses only have a display and a microphone, but no camera, although the 2 are exploring the opportunity of adding it to a future model.
Users still have to have their smartphones handy to assist power the glasses and get “real time info prompts and answers to questions,” per Nguyen. The glasses, that are manufactured by one other company that the startup didn’t name, are tethered to an accompanying app on the owner’s phone, where the glasses essentially outsource the computing since they don’t have enough power to do it on the device itself.
Under the hood, the smart glasses use Google’s Gemini and Perplexity as its chatbot engine, in accordance with the 2 co-founders. Gemini is healthier for math and reasoning, whereas they use Perplexity to scrape the web, they said.
During an interview, TechCrunch asked if their glasses knew when the following season of “The Witcher” would come out. Responding in a way paying homage to C-3PO, Ardayfio said: “‘The Witcher’ season 4 can be released on Netflix in 2025, but there’s no exact date yet. Most sources expect it within the second half of 2025.”
“I don’t know if that’s correct,” he added.
