The Empty Vessel
Apple is hosting a Welcome Home event in Tribeca on October 13, confirmed by Apple SVP Greg Joswiak on Oct. 8. The rumors around it are loud, led by Bloomberg’s Mark Gurman. We are looking at a roughly 6-inch hub (whether it ends up being called the HomePad or the HomeView is anyone’s guess) packed with an A18 chip and Face ID. It is a sleek piece of hardware that will likely set you back about $350 when it ships in mid-November. But there is a glaring problem: the screen ships, but the brain does not.
Hardware Ready, Intelligence Delayed
We are essentially being asked to buy a high-end display for a legacy assistant. While the A18 chip is more than capable of handling heavy lifting, the actual conversational agent control for HomeKit is nowhere to be found. Industry reports suggest we are waiting until late 2026 or early 2027 for that functionality. Apple is selling us a beautiful, expensive vessel for a brain that is currently on an extended vacation. It is a classic case of the smart display becoming the home agent’s face, but without the intelligence to back it up, it is just a glorified photo frame.
The Competition is Already Working
Contrast this with Meta’s Muse, which launched on September 8. Muse is already out there handling actual, tangible tasks like email, travel planning, health tracking, and shopping. It is an agent that does things. Meanwhile, Apple’s Siri AI is still stuck in personal-only mode, leaving users to wonder why they should invest in a hub that cannot even manage their smart home effectively.
Enterprise Governance vs. Consumer Silence
On the enterprise side, xAI’s GrokBot is reportedly GA, complete with audit trails, network controls, and SSO integration via Cursor. It is built for utility and governance. Apple, by comparison, seems to be struggling with the vocabulary of agents. They are selling hardware today for a promise that might not materialize until late 2026 or early 2027.
The Strategic Disconnect
This is the free dongle problem in AI agents writ large. Apple is betting that the hardware form factor will be enough to keep us hooked until the software catches up. But in a world where competitors are shipping functional, agentic tools, a shiny new hub that waits until late 2026 or early 2027 for actual intelligence feels less like a feature and more like a placeholder. We are buying the hardware, but we are still waiting for the product.
