Best of CES recap with Sunny Madra | E1880

Episode Summary

Executive Summary: The hosts assess AI announcements and products showcased around CES, focusing on whether purpose-built hardware or software integrations solve real user needs. They praise Rabbit’s connected-app concept but doubt its dedicated device is necessary, highlight promising senior-care and child-safety products, and examine the GPT Store’s disruption of apps. Their central view: businesses with proprietary data and strong user experiences may outcompete general AI platforms.

Main Topics: Rabbit AI device and agentic transactions: Rabbit combines a distinctive handheld device with a custom language model that connects to services such as Spotify, Gmail, Uber, and DoorDash. The hosts praise the app integrations and accessible price, but question whether the hardware offers enough advantage over a smartphone and whether users will trust AI to place orders. CES and AI-enabled consumer hardware: The hosts see CES as reinvigorated by AI, which lets hardware makers create sophisticated experiences without building full operating systems. They emphasize that a device must serve a clear purpose, rather than relying on novelty or form factor alone. AI companion for older adults: ElliQ is presented as a companion device intended to reduce senior loneliness and support habits, exercise, cognitive games, reminders, communication, and access to local resources. The hosts praise its breadth of use cases and considered hardware design. AI-supported smartphone safety for children: MMGuardian and Samsung’s offering uses AI to analyze phone activity and alert parents to risks such as cyberbullying, sexting, drugs, violence, predators, or suicidal ideation. The hosts value that it can preserve a normal smartphone experience while giving parents more insight. GPT Store and the changing app landscape: The GPT Store enables conversational interfaces to third-party services, with examples including Canva and AllTrails. The hosts see strong potential but warn that weak quality controls and limited customer ownership could undermine developers or cannibalize their existing apps and subscriptions. Specialized AI experiences versus general platforms: The discussion argues that service providers may have an advantage because they possess domain-specific data and understand user behavior. Examples include DoorDash-style personalized carts and Yelp search, where conversational AI could make existing products more useful.

Key Arguments: Rabbit’s combination of custom hardware and a model connected to users’ services is compelling, but the same experience may be possible on a phone, making the device’s distinct purpose uncertain. Reliable AI transactions require more than natural-language understanding: hallucinations, incorrect orders, and unclear responsibility between the model provider and service provider remain barriers to adoption. App developers may be better positioned than general-purpose AI platforms to deliver dependable experiences because they own the underlying data and understand how customers use their services. Purpose-built AI hardware can be valuable when it solves a specific problem, as with senior companionship or child safety; novelty alone is not sufficient product-market fit. The GPT Store creates a new distribution channel, but developers must consider whether it will bring customers, share revenue, or instead weaken their direct relationship with users and cannibalize their products. The hosts expect AI capabilities such as multimodality, stronger reasoning, reliability, and user customization to improve, raising the importance of integrating AI into existing products.

Data Points: Rabbit device price: $199 - The hosts describe the device as being offered at a relatively accessible price, while questioning whether that price covers hardware, shipping, and returns. Rabbit pre-orders: Approximately 10,000 initially; 30,000 subsequently - The hosts say Rabbit sold out, citing an initial pre-order figure that grew to 30,000. Siri launch date: October 4, 2011 - Used to illustrate how long consumers have waited for a dependable voice assistant that can interact effectively with apps and services. Custom GPTs in the GPT Store: 3 million - A host cites this as the reported number available around the store’s launch, while noting that quality varies. ElliQ feature examples: 12 - A host counts twelve distinct use cases or feature areas described for the senior companion, including reminders, games, exercise, and communication.

Pivotal Quotes: "why is everybody so afraid to consummate a transaction with an LLM?" — Jason Calacanis: Frames the discussion of why consumers and companies have been reluctant to let AI assistants complete purchases. "Is ordering your Uber going to be distinctly better because of the form factor? And I don't think the answer to that is yes." — Jason Calacanis: Summarizes his skepticism that Rabbit’s dedicated hardware provides a meaningful advantage over a smartphone. "If you have the data and if you can put the engineering effort in to get a model to understand your business, your data, and create a user experience, I think that will win right now before kind of the aggregators can." — Sandeep Madra: States the case for domain-specific companies building AI into their own products rather than ceding the experience to general platforms.

Implications: AI adoption will depend on trustworthy, useful experiences—not demos alone. Hardware needs a clear job, while app businesses should use their data to build AI features, protect direct customer relationships, and carefully assess the benefits and risks of third-party AI platforms.