OpenAI’s device push faces a bigger test than Apple’s suit

Apple’s lawsuit against OpenAI may have grabbed the headlines, but the bigger risk to the company’s hardware ambitions is simpler: convincing people they actually want an AI device at all.

Apple alleges that former employees Tang Tan, its ex-24-year hardware chief, and software engineer Chang Liu took trade secrets when they left to join OpenAI. According to the lawsuit, OpenAI “wrongfully took Apple’s secret and confidential information,” raising the prospect that Apple’s manufacturing relationships, product specs, and strategy could inform a future OpenAI device that might one day compete with the iPhone.

That legal fight lands just as OpenAI is trying to turn its hardware plans into something real. The company accelerated those efforts in 2025 with the acquisition of io Products, led by former Apple designer Jony Ive. Ive’s partnership with Sam Altman signaled high ambitions from the start, given his role in designing the iPhone, MacBook, and Apple Watch.

Reports have described OpenAI’s ideal product as a “third” device that would sit alongside a phone and a computer. The latest version of that idea is a portable, screen-free smart speaker with a “humanlike” personality. The lack of a screen appears to be central to the concept, and Ive has reportedly been wary of wearables such as pins, watches, and smart rings.

Other possible devices have also surfaced in reports, including a product rumored in January that would sit behind your ear for talking with ChatGPT, and a May report describing an “AI agent phone” powered by autonomous bots. But even if OpenAI builds a family of devices, it is still unclear whether any of them would feel meaningfully different from an Amazon Echo or Google Home.

Recommended reading

Avito eyes free dating inside its app

Why AI hardware keeps running into backlash

OpenAI also lacks the built-in commerce and smart-home ecosystems that Amazon and Google already have. If its devices are aimed mostly at work tasks, the audience may be narrow. The article points to the $230 limited edition Codex keypad as one example of hardware that may appeal mainly to hard-core AI users.

The financial appeal is obvious: hardware could open up a larger revenue stream than a $20-per-month ChatGPT subscription or inconsistent government contracts, especially as OpenAI continues to spend billions of dollars on development and hints at a future IPO.

But recent AI gadgets show how quickly things can go wrong. Meta’s smart glasses remain the best-known category leader, yet they have triggered repeated privacy concerns. This summer, Meta faced criticism after reports it was experimenting with facial recognition in the glasses. That was followed by a lawsuit alleging that Kenyan contractors could see what users saw through the glasses, including sensitive information and settings.

The Friend AI pendant drew similar criticism as an always-listening device. After the company spent $1 million on ads across New York City, some were quickly covered with graffiti reading “Surveillance capitalism” and “Get real friends.” Other products, such as the Plaud AI pin and Vocci AI smart ring, remain niche rather than essential.

That is the real challenge for OpenAI. Any device it releases will be judged not just against Apple’s claims, but against a growing list of AI hardware that has generated indifference, hostility, or privacy fears. For a company already asking users to trust it with powerful software, the usefulness of the product — and the credibility of its safeguards — may decide whether the category expands or stalls.