Solos AirGo V2: The Modular AI Smart Glasses — Transcript

Full transcript

1. 0:00So, today we are looking at the Solos
2. 0:02Airgo V2 today.
3. 0:03>> Mhm.
4. 0:04>> And this is a device that acts as an AI
5. 0:06companion that just well, it simply
6. 0:09looks at the real world with you.
7. 0:11>> The Airgo V2, it launched at CES for
8. 0:13around $299.
9. 0:15>> That price point is entirely dictated by
10. 0:17their physical design choices. They just
11. 0:19look like standard lightweight
12. 0:21eyeglasses.
13. 0:21>> They do. You can wear them to a cafe,
14. 0:24the office, the grocery store, and you
15. 0:27just blend in entirely. No one's going
16. 0:29to look twice at you.
17. 0:30>> But the underlying architecture making
18. 0:32that possible is what they're calling
19. 0:33the smart hinge system. I found the
20. 0:35mechanics of this so fascinating. The
21. 0:37frames and the internal electronics,
22. 0:39they aren't just glued together into one
23. 0:41monolithic piece of plastic. You can
24. 0:44actually pop the front frames off the
25. 0:45hinges and swap them out to match your
26. 0:47outfit, which is neat.
27. 0:48>> And more importantly, you can swap
28. 0:50between a frame that has a built-in
29. 0:52camera and a frame that's completely
30. 0:54blank.
31. 0:55>> Right, so you have options. Even the
32. 0:56temple arms, like the pieces that slide
33. 0:59over your ears, they clip right in and
34. 1:00out. So, I look at this device, I see
35. 1:02the modular frames, the slick design.
36. 1:05But if there is absolutely no augmented
37. 1:08reality display projecting onto the
38. 1:10lenses, can we legitimately call these
39. 1:13smart glasses?
40. 1:14>> That's the big question.
41. 1:15>> Right, like
42. 1:16>> I mean, that is the exact debate
43. 1:18happening in the tech community right
44. 1:19now.
45. 1:20>> Oh, really?
46. 1:20>> Yeah. And it's a totally valid critique
47. 1:22if you're strictly applying the old
48. 1:24definition of spatial computing.
49. 1:26>> Okay.
50. 1:26>> If smart requires a visual display, then
51. 1:29no, these aren't smart. But the
52. 1:30definition of ambient computing is
53. 1:32evolving.
54. 1:33>> Because rendering those graphics takes a
55. 1:35lot of juice.
56. 1:35>> Exactly. Rendering high-fidelity 3D
57. 1:38graphics in real time requires a massive
58. 1:41graphical processing unit.
59. 1:42>> And a massive battery.
60. 1:43>> Right, which generates a ton of heat
61. 1:45right on your face. So, Solos basically
62. 1:47sidesteps that entire thermal and power
63. 1:49nightmare.
64. 1:50>> Ah, I see. So, by stripping out the
65. 1:52screen, they free up all that processing
66. 1:54power and battery life to focus entirely
67. 1:57on sensory input and AI.
68. 2:00>> Precisely. The smart designation here
69. 2:02comes from real-world contextual
70. 2:05computing.
71. 2:06>> Meaning what, exactly?
72. 2:07>> Meaning it's not about projecting a
73. 2:09little digital weather widget into your
74. 2:12field of view, it's about the glasses
75. 2:14knowing you are staring at dark clouds
76. 2:16and answering you when you ask, "Hey, do
77. 2:18I need an umbrella today?"
78. 2:19>> Okay, that makes sense. Which means if
79. 2:21this AI is going to process the physical
80. 2:23world, it desperately needs to see.
81. 2:25>> In other eyes.
82. 2:26>> Right. And [snorts] the front frames
83. 2:28house this upgraded 16-megapixel
84. 2:31ultra-slim camera.
85. 2:32>> Which is no small thing.
86. 2:33>> Yeah, from a pure physics standpoint,
87. 2:35fitting a 16-megapixel sensor into a
88. 2:37standard eyeglass frame without, you
89. 2:39know, burning the user's face is a
90. 2:40serious engineering feat.
91. 2:42>> When you compress optical sensors and
92. 2:44image processors into a space that tiny,
93. 2:46they generate immense localized heat.
94. 2:48>> Right, they get super hot. Ow.
95. 2:50>> Yeah, wicking that thermal energy away
96. 2:52from the camera module and distributing
97. 2:55it along the frame so it just dissipates
98. 2:56harmlessly into the air.
99. 2:58>> That's wild. And it's not just taking
100. 3:00grainy still photos, either. We're
101. 3:01talking full HD video recording,
102. 3:03high-res photos, and even live video
103. 3:06streaming.
104. 3:06>> Yep.
105. 3:07>> It's essentially like a dashcam for your
106. 3:09life, but one that is directly wired
107. 3:11into a global encyclopedia.
108. 3:14>> A dashcam is a really brilliant way to
109. 3:15visualize it, but uh think about the
110. 3:18reality of
111. 3:19a camera mounted to human head for a
112. 3:21second.
113. 3:22>> Okay.
114. 3:22>> We do not move like cars. We constantly
115. 3:25micro-adjust our posture. We nod. We
116. 3:27dart our heads around to look at sudden
117. 3:29noises.
118. 3:29>> Oh, true. Raw video from a head mount is
119. 3:31usually pretty nauseating to watch.
120. 3:33>> Exactly. It looks like a chaotic found
121. 3:35footage horror movie.
122. 3:36>> Right, right. So, how do they stabilize
123. 3:38that for a live stream if you're walking
124. 3:39around with these on?
125. 3:40>> They utilize electronic image
126. 3:42stabilization or EIS.
127. 3:44>> EIS? Okay.
128. 3:46>> Right. So, instead of trying to
129. 3:47physically move the lens to compensate
130. 3:50for your head movement, which would
131. 3:52require bulky mechanical gimbals, they
132. 3:55do it digitally.
133. 3:56>> How does that work?
134. 3:57>> The camera captures a slightly wider
135. 3:59field of view than it actually needs.
136. 4:02Then, uses gyroscope data from the
137. 4:04glasses to understand exactly how your
138. 4:06head is moving.
139. 4:07>> Oh, so it tracks the tilt and panning?
140. 4:09>> Exactly. And it dynamically crops and
141. 4:12shifts the video frame in real time to
142. 4:15counteract your jitter.
143. 4:16>> Wow, okay.
144. 4:17>> The result is a surprisingly smooth,
145. 4:20really watchable video feed.
146. 4:22>> That makes the live streaming feature
147. 4:23genuinely usable then. But, I mean, that
148. 4:26brings us to the elephant in the room
149. 4:27whenever we talk about cameras mounted
150. 4:29to people's faces.
151. 4:30>> Privacy.
152. 4:31>> We have to talk about privacy.
153. 4:32>> People feel surveilled, the technology
154. 4:34will be rejected no matter how good the
155. 4:36AI is.
156. 4:37>> Which is why the inclusion of a highly
157. 4:38visible privacy indicator light on these
158. 4:41isn't just like a neat design choice.
159. 4:43>> No, it's vital.
160. 4:44>> It's a non-negotiable survival mechanism
161. 4:46for the product. Whenever the camera is
162. 4:49recording or streaming,
163. 4:50a distinct light on the frame turns on.
164. 4:53>> Yep, it illuminates bright enough so
165. 4:54everyone knows.
166. 4:55>> So, you cannot covertly film a
167. 4:57conversation. It immediately establishes
168. 4:59this baseline of trust in public spaces.
169. 5:02>> It forces accountability. So, the camera
170. 5:05acts as the sensory input, right? It
171. 5:06provides the raw data. But, raw data is
172. 5:08completely useless without
173. 5:10interpretation.
174. 5:11To understand how these glasses actually
175. 5:13function as an assistant, we have to
176. 5:15look at the brains behind the lenses.
177. 5:17>> Yes. The operating system they are using
178. 5:19is called Solos Chat 3.0.
179. 5:21>> Mhm.
180. 5:22>> And the defining feature of the system
181. 5:23is that it's a multimodal AI.
182. 5:25>> Multimodal, yeah.
183. 5:26>> Meaning, it isn't just waiting for you
184. 5:28to type out a text prompt. It is
185. 5:30simultaneously ingesting the live video
186. 5:33feed of the street you're on,
187. 5:35listening to the ambient audio of the
188. 5:37traffic, and processing your voice
189. 5:39commands all at once.
190. 5:40>> Which is incredible. Multimodal
191. 5:42capabilities are what bridge the gap
192. 5:44between like a simple search engine and
193. 5:47a genuine digital assistant.
194. 5:49>> Right.
195. 5:49>> But, what makes the Solos' approach
196. 5:52unique is that they aren't relying on a
197. 5:54single proprietary AI model.
198. 5:57>> Oh, right. They opened it up.
199. 5:58>> They did. They integrated four distinct
200. 6:00major platforms: ChatGPT, Claude,
201. 6:03Gemini, and DeepSeek.
202. 6:05>> Wait, wait. So, if I have all four of
203. 6:06those massive language models loaded in,
204. 6:09how does it know which one to use?
205. 6:10>> Well, you can choose.
206. 6:11>> Really? Can I essentially hot swap the
207. 6:13AI's brain depending on what specific
208. 6:16task I'm trying to accomplish?
209. 6:18>> can, and that is a massive strategic
210. 6:19advantage here.
211. 6:20>> Wow.
212. 6:21>> Because, you know, no single AI model is
213. 6:23the best at everything. They all have
214. 6:24their own distinct architectural
215. 6:26strengths.
216. 6:27>> Could you give me an example of that?
217. 6:28>> Sure. So, let's say you're looking at a
218. 6:29complex engineering schematic.
219. 6:32You might route that through Claude
220. 6:34because it has a stellar reputation for
221. 6:36structured logic and technical analysis.
222. 6:38>> Okay, that makes sense.
223. 6:39>> But, if you're traveling and need to
224. 6:40quickly identify a specific landmark in
225. 6:43real time, you might switch over to
226. 6:45Gemini since it's deeply integrated with
227. 6:48Google's live search index.
228. 6:50>> Oh, that is so smart. You're constantly
229. 6:52choosing the most specialized tool for
230. 6:53the job.
231. 6:54>> Exactly.
232. 6:55>> And you interact with these brains
233. 6:57entirely hands-free via a wake word.
234. 7:00But, there's a specific phrase in the
235. 7:02slide deck used to describe these
236. 7:03interactions that I think is critical.
237. 7:06>> Multi-turn natural conversations.
238. 7:09>> Yes, multi-turn.
239. 7:10>> That is the feature that prevents you
240. 7:11from wanting to throw the glasses into a
241. 7:13river after 10 minutes.
242. 7:15>> Right. Because with older voice
243. 7:16assistants, interaction was extremely
244. 7:19transactional.
245. 7:19>> It was exhausting.
246. 7:20>> to say the wake word, ask a question,
247. 7:22get your answer, and then say the wake
248. 7:24word all over again just to ask a
249. 7:26follow-up.
250. 7:26>> Yeah. Hey, assistant, what is this
251. 7:28building? It is the Louvre. Hey,
252. 7:29assistant, when was it built?
253. 7:31>> Exactly. It feels like you are
254. 7:32interrogating a robot rather than having
255. 7:35a dialogue.
256. 7:36>> Multi-turn capability changes that
257. 7:39entirely by utilizing what's called a
258. 7:41continuous context window.
259. 7:43>> Okay. How does that work?
260. 7:44>> So, the AI actively holds the memory of
261. 7:47your ongoing conversation in its
262. 7:48short-term token memory.
263. 7:50>> Right.
264. 7:50>> You ask, "What is this building?"
265. 7:52>> Mhm.
266. 7:52>> After it answers, you naturally just
267. 7:54follow up with, "Who designed it?" And
268. 7:56the AI inherently understands that it
269. 7:59refers to the Louvre.
270. 8:00>> That's huge. The interaction flows
271. 8:02exactly like a normal human
272. 8:03conversation.
273. 8:04>> Exactly.
274. 8:05>> Think about the practical applications
275. 8:06of that for a second. You're staring at
276. 8:08a menu in Tokyo, and it's translating
277. 8:10the text in real time.
278. 8:11>> Mhm.
279. 8:11>> Or, you know, you're at the hardware
280. 8:13store staring at a massive wall of
281. 8:14screws, totally lost.
282. 8:16>> You've all been there.
283. 8:17>> Right.
284. 8:17>> Yeah.
285. 8:17>> You just look at the wall and ask the
286. 8:18glasses, "Which one of these do I need
287. 8:20for half-inch drywall?" And it guides
288. 8:22you.
289. 8:22>> But, facilitating that kind of seamless
290. 8:25magic takes a massive toll on the
291. 8:27hardware.
292. 8:28>> I can imagine.
293. 8:29>> Having an ongoing, continuous dialogue
294. 8:32with a cloud-based AI while
295. 8:33simultaneously streaming video requires
296. 8:36two critical things,
297. 8:38>> which are?
298. 8:38>> a private way for the AI to speak back
299. 8:41to you, and an enormous amount of power.
300. 8:43>> Uh, yeah. That's the ultimate wearable
301. 8:45paradox, isn't it? You want infinite
302. 8:47battery life, but you don't want to wear
303. 8:49a heavy brick on your face.
304. 8:51>> Precisely.
305. 8:52>> Let's tackle the audio first, because if
306. 8:54I'm having a detailed conversation with
307. 8:56Claude about a private work email,
308. 8:58I really don't want my AI blasting its
309. 9:00responses for the entire subway car to
310. 9:02hear.
311. 9:03>> Yeah, that would be a deal-breaker.
312. 9:05But, Solo solved this using directional
313. 9:07open-ear speakers built directly into
314. 9:09the temple arms.
315. 9:10>> So, there's no physical earbud jamming
316. 9:12into your ear canal.
317. 9:13>> Right, nothing in your ear.
318. 9:15>> Which is fantastic for situational
319. 9:16awareness. You can still hear traffic or
320. 9:18someone calling your name.
321. 9:19But, how does an open speaker near your
322. 9:22ear remain private?
323. 9:24>> It's pretty clever.
324. 9:25>> Like, why doesn't the person sitting
325. 9:26next to me hear it?
326. 9:27>> It relies on a principle called phase
327. 9:29cancellation.
328. 9:30>> Okay, I've heard of that with
329. 9:31noise-canceling headphones.
330. 9:32>> Similar concept, yeah. Instead of just
331. 9:35pushing sound waves out in all
332. 9:36directions like a traditional speaker,
333. 9:38these arms use an array of micro
334. 9:40speakers firing at precisely timed
335. 9:43intervals.
336. 9:44>> Okay.
337. 9:45>> Imagine dropping two pebbles into a pond
338. 9:48where the ripples meet, they either
339. 9:49amplify each other or they completely
340. 9:51cancel each other out.
341. 9:52>> Right, destructive interference.
342. 9:54>> Exactly. These speakers essentially
343. 9:56construct a targeted acoustic bubble
344. 9:59right over your ear canal.
345. 10:01>> That is so cool.
346. 10:02>> For you, the sound waves arrive
347. 10:03perfectly in sync, creating crisp, clear
348. 10:06audio. But for the person sitting two
349. 10:08feet away, those same sound waves arrive
350. 10:11out of phase and literally erase each
351. 10:13other in the air before reaching their
352. 10:15eardrum.
353. 10:15>> Wow, that is brilliant engineering. But
354. 10:18okay, processing phase array audio,
355. 10:21capturing stabilized video, pinging four
356. 10:23different AI servers. The battery drain
357. 10:26has to be astronomical.
358. 10:28>> Oh, it is.
359. 10:28>> Like we're talking about a recipe for a
360. 10:29dead device in 45 minutes.
361. 10:31>> laws of physics dictate that battery
362. 10:33capacity requires physical volume. You
363. 10:35cannot cheat the chemistry. But they
364. 10:37engineered an incredibly clever
365. 10:39mechanical workaround using that smart
366. 10:42hinge system we talked about earlier.
367. 10:44>> The battery temples.
368. 10:45>> Yes.
369. 10:45>> This is honestly my favorite hardware
370. 10:47design choice on the whole device.
371. 10:49>> It's so smart.
372. 10:50>> arms out.
373. 10:52The pieces resting on your ears actually
374. 10:53are the batteries.
375. 10:54>> Yes.
376. 10:55>> And because they're hot swappable, you
377. 10:56don't have to take the glasses off,
378. 10:58power them down, and tether yourself to
379. 11:00a wall charger when they die.
380. 11:01>> No cords required.
381. 11:02>> You literally just pop off the dead arm
382. 11:05and click on a fully charged one. It's
383. 11:07like reloading a high-tech gadget in a
384. 11:09spy movie.
385. 11:09>> That's a great comparison.
386. 11:11>> Click out the old clip, snap in a fresh
387. 11:12one, and you're immediately back in the
388. 11:14action.
389. 11:15>> It bypasses the physical limitations of
390. 11:17battery size entirely, and the base
391. 11:20battery life on a single set of arms is
392. 11:22highly respectable.
393. 11:23>> What are they quoting?
394. 11:24>> They're quoting about 10 hours of
395. 11:26continuous music playback or 7 hours of
396. 11:29active call time.
397. 11:30>> Okay, wow. Which covers a full work day
398. 11:32for most people.
399. 11:33>> Easily. And for heavy power users, they
400. 11:36designed an optional charging case.
401. 11:38>> AirPods case.
402. 11:39>> Sort of.
403. 11:40But it looks identical to a standard
404. 11:43premium eyeglasses case.
405. 11:45Inside though, it houses an 1,100
406. 11:47milliamp hour battery.
407. 11:49>> Oh, nice.
408. 11:50>> So you can be actively wearing the
409. 11:52glasses and talking to the AI while your
410. 11:54spare set of battery arms is sitting in
411. 11:56the case inside your bag recharging.
412. 11:59>> That's amazing.
413. 12:00>> Yeah.
414. 12:01>> built in an auto power on and off
415. 12:02feature, right? You take the glasses
416. 12:04off, you set them on your desk, and the
417. 12:06internal gyroscopes realize they aren't
418. 12:08being worn, so they instantly put them
419. 12:10to sleep to conserve juice.
420. 12:11>> And you put them back on, they wake up.
421. 12:13>> It's all designed to completely remove
422. 12:15the friction of interacting with the
423. 12:16technology.
424. 12:17>> But you know, power is only half of the
425. 12:19equation here.
426. 12:20>> What's the other half?
427. 12:20>> The other invisible tether we need to
428. 12:22discuss is data.
429. 12:23>> Ah.
430. 12:24>> Processing high-res video and sending it
431. 12:27to an AI in the cloud requires moving a
432. 12:29massive amount of data extremely fast.
433. 12:31>> Right, because to keep the glasses
434. 12:33lightweight and affordable, the actual
435. 12:35heavy computing isn't happening inside
436. 12:36the frames.
437. 12:37>> No, there's no room for that processor.
438. 12:39>> Your smartphone is acting as the
439. 12:40middleman, pulling the data from the
440. 12:42glasses and routing it to the cloud.
441. 12:43>> Exactly.
442. 12:44>> And they use a dual connectivity system
443. 12:46to manage this traffic jam, don't they?
444. 12:48>> Yeah, it's a very elegant networking
445. 12:50solution.
446. 12:51For the low-intensity tasks like
447. 12:53connecting to the companion app, taking
448. 12:55phone calls, streaming a podcast, or
449. 12:57just basic voice commands, the glasses
450. 13:00use standard Bluetooth.
451. 13:02>> Okay.
452. 13:02>> Think of Bluetooth as a quiet, narrow
453. 13:04country lane. It requires very little
454. 13:07power, making it incredibly energy
455. 13:09efficient, which is perfect for a slow
456. 13:11trickle of audio data.
457. 13:12>> But a live 1080p video feed is not a
458. 13:15slow trickle.
459. 13:15>> Not at all.
460. 13:16>> It is a massive convoy of heavy freight
461. 13:18trucks.
462. 13:19>> Yeah.
463. 13:19>> If you try to force that much data down
464. 13:22a narrow dirt path,
465. 13:23you get massive traffic jams. You get
466. 13:25latency.
467. 13:26>> And latency kills the experience.
468. 13:28>> If the AI is trying to identify a
469. 13:29fast-moving object, even a 1-second lag
470. 13:33makes the entire system useless.
471. 13:34>> Exactly. So, to circumvent that
472. 13:36bottleneck, the glasses also include a
473. 13:39low-power Wi-Fi connection.
474. 13:41>> Oh, built right in?
475. 13:42>> Yes. The millisecond you activate a
476. 13:44high-bandwidth task, like live streaming
477. 13:47or asking the AI to analyze a live
478. 13:49visual feed, the glasses dynamically
479. 13:52spin up a direct Wi-Fi connection to
480. 13:54your phone.
481. 13:55>> Wow.
482. 13:56>> They essentially build a dedicated
483. 13:58multi-lane high-speed highway to
484. 14:00guarantee the data transfers instantly
485. 14:03without lag.
486. 14:04>> This seamless. And all of this routing,
487. 14:06the model selection, the device
488. 14:08management, it all happens through the
489. 14:10Solos Airgo companion app.
490. 14:12>> Right. It is entirely dependent on your
491. 14:13phone.
492. 14:14>> But this connectivity pipeline opens the
493. 14:16door to something far more profound than
494. 14:18just translating menus or finding
495. 14:19drywall screws.
496. 14:20>> definitely.
497. 14:21>> Solos announced a major integration with
498. 14:23Envision, a company that specializes in
499. 14:25assistive technology.
500. 14:26>> Which is huge.
501. 14:27>> We've been talking about this as a
502. 14:29productivity hack. But with Envision,
503. 14:31they are planning tools like dedicated
504. 14:33scene description, text recognition, and
505. 14:36object identification designed
506. 14:38specifically to help blind or low-vision
507. 14:41users navigate the physical world.
508. 14:43>> I'd argue it's even more fundamental
509. 14:45than a productivity hack. It's a
510. 14:47paradigm shift in independence.
511. 14:49>> Truly. A visually impaired user puts
512. 14:51these glasses on and the AI can read the
513. 14:54fine print on a prescription medicine
514. 14:56bottle to them.
515. 14:57>> Right. Or it can verbally announce what
516. 14:59number bus just pulled up to the transit
517. 15:01stop.
518. 15:01>> a consumer tech gadget into a literal
519. 15:04seeing eye assistant.
520. 15:05>> And historically, specialized medical
521. 15:07devices that perform those exact
522. 15:09functions cost thousands of dollars.
523. 15:11>> And they often look highly clinical,
524. 15:12which carries its own social stigma.
525. 15:14>> Exactly. Packing that life-changing
526. 15:17utility into a $299 device that just
527. 15:20looks like standard eyewear is
528. 15:22revolutionary.
529. 15:23>> And they're actively encouraging the
530. 15:25tech community to push this even
531. 15:26further.
532. 15:27>> Yeah, that's DK.
533. 15:28>> Right.
534. 15:29Solos released an Ergo SDK, a software
535. 15:32development kit, meaning they are
536. 15:34opening up the underlying architecture
537. 15:36so third-party developers can build
538. 15:37their own custom applications.
539. 15:39>> Because hardware is ultimately only as
540. 15:41valuable as the software ecosystem that
541. 15:43supports it.
542. 15:44>> Right.
543. 15:44>> By releasing an SDK, Solos is signaling
544. 15:47that they don't just want to sell a
545. 15:48standalone product. They want to
546. 15:50establish a platform.
547. 15:52>> So anyone can build for it.
548. 15:53>> Yeah. Developers can now build highly
549. 15:56specialized enterprise workflows.
550. 15:58Imagine an application built for
551. 15:59warehouse logistics where workers can
552. 16:01identify inventory and cross-reference
553. 16:03databases entirely hands-free.
554. 16:05>> Oh, that would be incredibly efficient.
555. 16:07>> Or specialized navigation apps for
556. 16:09university campuses.
557. 16:10>> Looking at the reality of what the Ergo
558. 16:13V2 offers,
559. 16:14does that $299 price tag actually hold
560. 16:18up? Let's weigh the pros and the cons.
561. 16:20>> I mean, the value proposition here is
562. 16:22remarkably strong on the pro side.
563. 16:24>> Agreed.
564. 16:25>> You're gaining immediate hands-free
565. 16:27access to the most powerful multimodal
566. 16:30AI models on the planet, OpenAI,
567. 16:32Anthropic, Google.
568. 16:34>> Plus the hardware itself.
569. 16:35>> Right. You're getting a highly modular,
570. 16:37repairable hardware design with that
571. 16:39brilliant hot-swappable battery system.
572. 16:42The camera provides stabilized,
573. 16:44genuinely usable footage.
574. 16:46>> And achieving all of that at a $299
575. 16:48price point makes contextual computing
576. 16:50accessible to a massive demographic.
577. 16:52>> It really does. But the limitations are
578. 16:54equally real, and they're inherently
579. 16:56baked into the philosophy of the device.
580. 16:58>> glaring con for power users is going to
581. 17:02be the absolute lack of an augmented
582. 17:04reality display.
583. 17:05>> Yeah, that's a hard stop for some
584. 17:06people.
585. 17:07>> If your primary goal is having digital
586. 17:09maps overlay directly onto the pavement
587. 17:11or seeing your text messages floated
588. 17:13midair, this device fundamentally cannot
589. 17:15do that.
590. 17:16>> Second is the severe dependency on your
591. 17:19smartphone.
592. 17:20>> Right, it's not a standalone unit.
593. 17:21>> No, these glasses are brilliant sensory
594. 17:23inputs, but they are not standalone
595. 17:26computers. Your phone is doing all the
596. 17:28heavy lifting, routing the data, pinging
597. 17:30the cloud.
598. 17:31>> Which means this system is going to
599. 17:33aggressively drain your phone's battery
600. 17:36alongside its own.
601. 17:37>> Oh, absolutely.
602. 17:38>> And finally, the social friction.
603. 17:40Despite the inclusion of that very
604. 17:42necessary privacy indicator light, the
605. 17:44cultural unease of wearing a camera on
606. 17:46your face in public spaces is going to
607. 17:48linger.
608. 17:49>> It definitely will. It'll take years of
609. 17:51normalized use for that discomfort to
610. 17:53fully fade. It is a deliberate,
611. 17:55calculated trade-off. Solos looked at
612. 17:57the landscape and decided that a
613. 17:58lightweight, affordable, AI-first
614. 18:01wearable was a much faster path to mass
615. 18:03adoption than a heavy, incredibly
616. 18:06expensive, visually focused headset.