AI the Product vs AI the Feature: A Case Study of Humane AI Pin and Rabbit R1

print("AI the Product vs AI the Feature: A Case Study of Humane AI Pin and Rabbit R1")

This is the 20th consecutive post on this AI Series journey! 🎉🦾🙂.

A perfect time for a mid-series special 🤗.

The rise of Artificial Intelligence(AI) has been at full speed since the success of ChatGPT which was launched on November 30, 2022, a profound moment in the history of AI and modern technology. From then on everyone wanted a piece of that magic pie, VCs, startups, Fortune 500 companies, etc... everyone wanted to "integrate" AI into their product and try to say their product "uses AI" 🙂.

We have seen a lot of these products ranging from plain software products like; Open Art for image generation, Udio for music generation, Synthesia for video generation, etc. To hardware and software products like the Rabbit R1 and the Humane AI Pin.

It is a gold rush period for AI right now. But these products have had varying levels of success to morbidly bad failures. That begets the question, is AI a product or a feature? Is it more useful as a feature or product? Let's find out together 👽.

First of all, let's be clear on what a feature and a product are.

A product can be described as something that customers consider valuable enough to pay for. It is an item, tool, environment, or even a service for which buyers would be willing to pay.

A feature on the other hand is a part of the product.

So for example your phone is a product, and the ability to make calls and connect to the internet are features of that phone. Another example is an app say YouTube, while Youtube itself is the product the features are comments, likes, playlists, etc.

With what we now know about a product and a feature, we can safely conclude that a product is a collection of 1 or more features 🤖.

Now that we are clear on what a product and a feature are, let's compare AI the product vs AI the feature.

AI the product

We have seen the rise of very successful AI products like ChatGPT, Synthesia, Perplexity AI, etc. On the other hand, we have also seen terrible AI products like the DOA products Humane AI Pin and Rabbit R1. So what, some products succeed and some fail right? Or is this scenario different? Well, let's check AI the feature to better understand our hypothesis.

AI the feature

AI the feature can be seen in products integrating features like Whatsapp adding Meta AI which is basically like ChatGPT, Apple's next IOS, IOS 18's email summary feature, and Siri upgrade which basically killed products like Humane AI Pin and Rabbit R1. There are a lot of these examples, but let me add one more, GitHub Co-Pilot which basically also integrates custom ChatGPT to Visual Studio Code as a feature.

From our understanding of AI the product and the feature, the question to ask now is this, which is better?

Let us use an analogy with the social network Clubhouse, a real-time virtual "room" in which users can communicate with each other via audio. Now Clubhouse got really popular during the 2020 pandemic with everyone "sitting at home". It got so successful that other products copied it and added it as a feature, like Slack's Huddle and Twitter Spaces. At its peak during the pandemic, Clubhouse had about 10 million monthly downloads, but barely 1 year later it barely makes up to a 1k.

The point of the Clubhouse illustration is to show that Clubhouse as a feature was better and more successful than Clubhouse as a product.

To buttress our point let us look at the case of some of the worst-received AI products.

The case of Humane AI Pin and Rabbit R1

When I first saw the Rabbit R1 Introduction video I was really impressed, you can check it out here. But at the same time, I felt like, was this enough to build a whole hardware for? Would it not be better as an app like ChatGPT? It turns out I was right. Also, I believed that most of the things Rabbit mentioned as key features to differentiate them from other chatbots like their Large Action Model were features that can be added to Siri or Google Assistant.

The Large Action Model (LAM) was supposed to be able to interact with an app and across multiple apps from voice commands. It was supposed to be able to do things like schedule flights and vacations, choose restaurants, and make reservations in seconds by interacting across the different apps needed to achieve that. Sounds promising right?

As it turns out I was right, not just because of my initial thoughts, but because the product was simply terrible it could not do any of the LAM features it boasted of, and even the simple things, it felt rushed and there was a scenario where it was reported recently that hardcoded environment variables were found in their codebase when it was reverse engineered, you can watch that here.

This brings us to the Humane AI Pin. To be sincere I had no idea what problem the product was solving, even after watching the introduction video and even after many reviews by top tech reviewers. You can watch the introduction video here. Humane AI apparently built a new smart device based on AI and without apps, with new ways of interacting with the device by hand gestures but primarily by voice. It was not really offering anything more like the Rabbit R1 did with their LAM claims, it was basically a device with one feature, a voice assistant. So it didn't come as a surprise to me when the top tech reviewer MKBHD labeled it the "The worst product I have ever reviewed... for now", you can watch that here as well.

Conclusion

The big tech companies Apple, Meta, Samsung, etc see AI as a feature that can optimize and complement their products.

What if Rabbit R1 did everything it said in its introduction video? Would it have been successful? I believe maybe temporarily, but eventually, it would be better on a product like a smartphone using IOS or Android.

Would these AI products be better as features rather than standalone products? I believe yes 👽. We have seen companies adding AI features to their products even when not necessary and it's been fine, we have also seen how difficult it's been for AI products.

Whether we are right or wrong time will tell... but for now, I believe AI the feature is best!

The end.

How was this chapter? Let me know what you think about AI the feature and the product.

Have a nice time, we'll continue with an awesome chapter again next week.

Take care 👽

⬅️ Previous Chapter

Next Chapter ➡️