This week, startup Humane seeded its long-awaited seed. AI pin We appealed to reviewers and influencers, but the response was overwhelmingly low. Most notable was a review by Marques Brownlee (aka MKBHD), who has 18 million followers on YouTube and 6.2 million followers on “The worst product ever.” Condition… This stuff is basically always bad in almost everything. ”
Related article: Humane Ai Pin review: 5 points about a promising but dangerously flawed wearable
The main complaints from Brownlee and other reviews focus on the practicality and performance of the device rather than the hardware itself (the pins seem well made). There are a few things that AI pins fall short of.
- Concept and execution: AI Pin is an innovative wearable AI assistant that works independently and is equipped with high-quality hardware. However, there are significant functionality issues, including slow and inaccurate responses.
- User interface issues: The device's new gesture-based controls and Class 2 monochrome green laser projector interface are not easy to use in a variety of situations, require a steep learning curve, and are difficult to see in bright environments.
- Reliability concerns: AI pins have short battery life and overheating issues, requiring frequent charging and causing discomfort.
- Limited compatibility: The device lacks integration with smartphones and popular apps, limiting its practicality despite its standalone nature.
- Cost and value: Given its limited functionality compared to modern smartphones, it's hard to justify AI Pin's high price and additional subscription fees.
- Future possibilities and current limitations: Although aimed at reducing screen time, AI Pin's current limitations make it unattractive compared to the overall functionality of a smartphone.
It is unclear whether Humane will be able to improve the performance and reliability of current devices through embedded firmware updates and backend software development. Certain aspects of the hardware (such as the difficulty of using laser projectors under varying lighting conditions) may limit performance, and other constraints may limit only small incremental improvements. .
Battery charging and overheating issues may also not be resolved with current hardware.
connection issues
However, there may be some things Humane can do. Let's start with connection response time.
This device uses 4G LTE data provided by T-Mobile. This is fine in well-connected areas, but if you don't have coverage like any other mobile device, it becomes a roaming issue (Humane and T-Mobile contracts allow roaming) case). Not all areas have good wireless coverage across multiple providers.
Also, what is Wi-Fi 7 and how fast is it?
Another issue is that if you are in a very congested area, 4G network access speeds will be slow and your connection will drop. This is one of the reasons why carriers have been investing heavily in the rollout of 5G to address congestion issues in urban areas.
Wearables require a reliable data connection to function properly. Humane Pin offers dual frequency Wi-Fi with both 2.4Ghz and 5Ghz antennas. However, it is of little use due to the limited number of compatible Wi-Fi networks. This device has a hard time authenticating on public Wi-Fi networks, including most networks outside your home, such as airports, airplanes, cafes, restaurants, and workplaces, such as networks that require a web portal for authentication.
Even a simple password-type connection to a broadcast SSID can be a pain. Humane documentation states that joining a Wi-Fi network requires voice and laser UX intervention, or a QR code (which must be printed so that it can be recognized by AI Pin's camera) .
Also: The best hotspots you can buy
Oh, and T-Mobile's public hotspots? Good luck. they don't exist. AT&T and Verizon have these services, as do many broadband providers such as Comcast Xfinity… pin is not accessible.
Can I use my 5G smartphone as a hotspot? Yes, you can. But that defeats the purpose of Humane paying $25 a month for his 4G service.
In terms of independent data connectivity, if the Humane Pin can use an eSIM from any provider (assuming its 4G LTE radio works with a US provider, like the iPhone and most Android smartphones), you'll pay regular rates. If you can, it will be more reasonable. The smartwatch costs $15 per month and can be sent to your preferred carrier.
Of course, the $25 fee also includes cloud storage for videos and photos, GenAI computing, and other infrastructure costs. However, if your device is Wi-Fi only or you can use your existing provider, it can cost at least $10.
Related article: Apple Watch Series 9 review: Why you're not satisfied with the cheaper model
To resolve Wi-Fi issues, the device requires a pass-through form of authentication, similar to how the iPhone works with the Apple Watch, over Bluetooth to your smartphone. Of course, this will require the development of a smartphone app. At the very least, you should have a minimal app to avoid connectivity issues.
My iPhone can store many Wi-Fi networks (that I have authorized) in memory. It uses Bluetooth to pass them to the Apple Watch, and it's completely invisible to me. That's exactly what the pin needs.
4G and Wi-Fi aren’t the only answers for wearable connectivity
However, Wi-Fi consumes a lot of power even when implementing low-power transceivers. A possible solution for future Pin hardware is to incorporate Matter, an IoT connectivity standard adopted by many vendors including Apple, Amazon, Google, and Samsung. Matter provides a universal language for wireless (and wired) IoT devices, allowing smart home products from different manufacturers to work together.
Also: Everything you need to build a smart home
Wireless Matter devices can connect to each other using the Zigbee wireless protocol. Smart home bridges such as Amazon Echo, Amazon eero, Philips Hue, Apple TV 4K, Apple HomePod, and other devices with Matter and Zigbee built-in can be used for this purpose. Zigbee supports data rates of up to 250 kbps on 2.4 GHz and 40 kbps on 915 MHz in North America, with a range of up to 100 meters. Zigbee's repeater and mesh networking capabilities can significantly extend its signal range. Amazon sidewalk is an example of a long-range mesh network that uses 915 MHz Zigbee and Bluetooth Low Energy, a combination that can also be used with PIN.
Matter's data transmission speeds may not be as fast as Wi-Fi for moving large amounts of data (such as moving photos and videos to cloud storage), but it can handle text-based LLM queries and sensor data on wearable devices. It serves as a side-channel communication mechanism. , it makes a lot of sense.
LLM and cloud provisioning dilemma
However, not all of your device's slowness is due to the connection. Query responses from Humane's AI are also said to be slow and unreliable, likely due to under-provisioning of cloud resources. Ping's backend GenAI LLM is clearly OpenAI. I'm not sure whether to use GPT4 or GPT 3.5, but it should be done via an API call.
This pin is unlikely to make a direct API call to OpenAI on Azure. Queries are likely to be proxied and rapidly modified by cloud infrastructure running on any number of providers, including Azure OpenAI (very likely), AWS, and Google Cloud. will be higher. Depending on how well it is provisioned, it contributes to the overall performance of the service.
Also: GPT-4 Turbo takes back the crown of “Best AI Model” from Anthropic’s Claude 3
Here's a question: Is it wise to limit pins to OpenAI/ChatGPT? Other LLMs are currently in development, including Meta's LLAMA2, Google Gemini, and Amazon Titan. Advances in AI are rapid and furious, with model parameter sizes increasing multiple times a year. Competition between these models is fierce. AI-driven products that are not limited to a particular ecosystem should be service agnostic and always have access to the best model and user-preferred LLM.
Can the pin be fixed, or will rivals take advantage of Humane's failure?
Standing at the intersection of innovation and practicality, Humane's AI Pin marks a bold foray into the future of wearable technology. But the path from breakthrough concept to essential device will be difficult. Issues with connectivity, user interface, and overall utility highlight the disconnect between Humane's visionary ambitions and the practical expectations of its user base.
To close this gap, humanity urgently needs to enhance the performance of AI pins. Enhancements include improved connectivity with updates such as eSIM support and more sophisticated WiFi solutions. Make the user interface more accessible. Make your cloud services more resilient. Adopting IoT standards like Matter could further integrate AI Pin into the broader digital ecosystem, enhancing its versatility and appeal.
Also, Limitless' $99 AI wearable promises to remember your meetings and everything else
However, Humane does not operate in a vacuum. Tech giants like Apple, Amazon, and Google, with their vast resources, established device ecosystems, and advanced GenAI cloud capabilities, are well-positioned to take advantage of Humane's stumbling blocks. These technology giants have the expertise and infrastructure to quickly implement similar innovations, and may be more successful in refining and integrating wearable AI technology into their product lines. With its established user base and robust cloud offering, it has the potential to outperform Humane by offering a more sophisticated and user-friendly alternative that more closely aligns with consumer expectations.
As the personal technology landscape continues to evolve, Humane is under pressure to rapidly iterate and improve its AI pins. If you can effectively respond to user feedback and improve your product, you still have a chance to compete in this dynamic market. Otherwise, it risks being overshadowed as competitors like Apple, Amazon, and Google leverage their technological prowess to capture the imagination and trust of consumers looking for the next big thing in wearable technology. .