OpenAI and former Apple design chief Jony Ive have reportedly been toiling for much of the year on a clandestine artificial intelligence device. Initial whispers hinted at a launch by late 2026, but the project now appears to be navigating a sea of technical difficulties. A recent report indicates that the teams are wrestling with substantial software and infrastructure obstacles, specifically regarding how to secure and manage the immense computing resources required to operate OpenAI’s sophisticated AI models on a large scale. This critical challenge could significantly impede progress on the much-anticipated device.
OpenAI’s AI Device Reportedly Faces Multiple Issues
Sources suggest that while the physical development of this secretive AI gadget is proceeding without a hitch, the company is grappling with key software elements and the underlying infrastructure. Insiders claim that a major point of contention on the software side is defining the AI assistant’s ‘personality’. Data privacy is also highlighted as a significant concern.
From an infrastructure standpoint, managing the compute budget stands out as a colossal challenge OpenAI is striving to conquer. This very issue is believed to be the primary factor that could push back the device’s debut. One source, speaking to Financial Times, articulated the problem bluntly: “Amazon has the compute for an Alexa, so does Google [for its Home device], but OpenAI is struggling to get enough compute for ChatGPT, let alone an AI device; they need to fix that first.”
Regarding the device’s form, reports describe it as roughly the size of a smartphone, allowing users to interact through integrated cameras, microphones, and speakers. It is said to incorporate multiple cameras and is designed primarily for desktop use, though it can also be carried by the user.
It reportedly remains “always on,” eliminating the need for a trigger word. Instead, it continuously gathers environmental data to assist the user throughout their day. This concept brings to mind similar offerings like Humane’s AI Pin or the Rabbit R1, but distinctively as a desktop solution rather than a wearable.
The device is reportedly envisioned as a significant leap beyond existing products like Amazon’s Echo and Google Nest, which offer assistants capable of only a predefined set of actions. OpenAI and Ive are reportedly aiming to create a more powerful, general-purpose AI device.
This ambitious goal is precisely where the software quandaries emerge. The development teams are reportedly struggling to refine the AI assistant’s voice and mannerisms. While they desire a versatile assistant suitable for various tasks, there’s a clear intention to avoid it sounding like “your weird AI girlfriend,” a critique sometimes leveled at Apple’s Siri. Another delicate balance they seek to strike is making the device genuinely helpful without becoming overly intrusive.