A decade after one of the most spectacular hardware failures in consumer electronics history, Amazon is quietly building another smartphone. This time, the company thinks artificial intelligence changes everything.
The effort, internally code-named “Project Transformer,” aims to create an AI-centric mobile device that would serve as a persistent companion powered by Amazon’s Alexa+ assistant, according to a report from Android Central, which cited details originally surfaced by journalist Ed Zitron. The device reportedly wouldn’t just run apps the way a conventional smartphone does. Instead, it would act as an AI-first interface — one where Alexa+ handles tasks, anticipates needs, and operates as the primary way users interact with the phone.
If this sounds familiar, it should.
Amazon’s Fire Phone, launched in June 2014 with considerable fanfare, was a catastrophic misfire. The device debuted at $199 on a two-year AT&T contract and featured gimmicky hardware tricks like Dynamic Perspective — a 3D display effect powered by four front-facing cameras that tracked the user’s head. It had Firefly, a feature that could identify products, music, and TV shows. The phone was, in essence, a shopping portal disguised as a smartphone. Consumers saw through it immediately. Within months, Amazon slashed the price to 99 cents. The company eventually took a $170 million write-down on unsold inventory, and CEO Jeff Bezos later acknowledged the failure publicly. The Fire Phone became shorthand in Silicon Valley for corporate hubris in hardware.
So why try again?
The answer lies in how dramatically the competitive dynamics of mobile computing have shifted since 2014. Back then, Amazon was trying to compete with Apple and Samsung on their own terms — spec sheets, app stores, camera quality. That was a losing proposition for a company whose core competencies were cloud infrastructure, logistics, and retail. But the emergence of large language models and agentic AI has created an opening that didn’t exist before. The smartphone, as currently conceived by Apple and Google, is essentially a grid of app icons that hasn’t changed meaningfully in 17 years. Amazon appears to be betting that AI can finally break that model.
Project Transformer reportedly runs on Android, which immediately solves one of the Fire Phone’s most crippling problems: app availability. The Fire Phone ran Fire OS, Amazon’s forked version of Android that lacked access to the Google Play Store. Users couldn’t get Gmail, Google Maps, YouTube, or thousands of other apps they depended on daily. It was a dealbreaker. By building on standard Android this time, Amazon would give users access to the full library of apps while layering its own AI-driven interface on top. Think of it less as a replacement for Android and more as an AI skin that changes how you use it.
The Alexa+ assistant sits at the center of the concept. Amazon launched Alexa+ earlier this year as a more capable, conversational version of its voice assistant, powered by large language models. It can handle multi-step tasks, maintain context across conversations, and integrate with third-party services more fluidly than the original Alexa. On a smartphone, Alexa+ could theoretically manage your calendar, order groceries, control smart home devices, book reservations, and respond to messages — all through natural language rather than tapping through individual apps.
That’s the pitch, anyway.
Amazon isn’t the only company chasing this vision. The past 18 months have seen a rush of AI-first hardware attempts, and the results have been sobering. The Humane AI Pin, which promised to replace smartphones with a screenless, voice-and-projection-based wearable, launched to devastating reviews and sluggish sales. The company reportedly explored a sale. The Rabbit R1, a bright orange handheld AI device, generated viral preorder interest but disappointed users with limited functionality and reliability issues. Both products shared a fundamental problem: they asked consumers to abandon their smartphones entirely. People weren’t willing to do that.
Amazon seems to have learned from these failures — and from its own. Rather than asking users to give up the smartphone form factor, Project Transformer reportedly embraces it. You still get a screen. You still get apps. You still get a camera. But the AI layer on top is supposed to make all of those things work together more intelligently. It’s an additive approach rather than a subtractive one, and that distinction matters enormously in consumer hardware.
The timing is also significant. Apple has been integrating its own AI capabilities under the Apple Intelligence banner, rolling out features like writing tools, image generation, and a more capable Siri across iPhones, iPads, and Macs. Google has been pushing Gemini, its multimodal AI model, deeper into Android and Pixel devices. Samsung has partnered with Google to bring Galaxy AI features to its flagship phones. The major players are all converging on the same thesis: AI will be the next interface layer for mobile devices. Amazon, despite its Fire Phone humiliation, clearly doesn’t want to cede this ground.
And it has assets the 2014 version of the company didn’t fully possess. Amazon Web Services is the world’s largest cloud computing platform and hosts the Bedrock service, which gives developers access to foundation models from Anthropic, Meta, Mistral, and Amazon’s own Nova family. The company has invested billions in Anthropic, the maker of Claude. It has a massive logistics and retail operation that an AI phone could tap into directly — imagine a device that knows what you need before you order it, or that can coordinate same-day delivery through voice alone. Amazon also has one of the largest installed bases of smart home devices through its Echo line, Ring doorbells, and Eero routers. A phone that acts as the central command node for all of those devices has a compelling value proposition that Apple and Google can’t easily replicate.
But compelling value propositions don’t guarantee market success. The smartphone market is brutally consolidated. Apple and Samsung together control roughly 55% of global smartphone shipments, and their dominance in the premium tier — where margins actually exist — is even more pronounced. Chinese manufacturers like Xiaomi, Oppo, and Vivo have locked up much of the mid-range and budget segments. Breaking into this market requires either a radically differentiated product or a willingness to subsidize hardware losses for years while building a user base. Amazon has shown willingness to do the latter with its Kindle and Fire tablet lines, selling hardware at or below cost to drive engagement with its services. A similar strategy could apply here.
The report from Android Central notes that details remain sparse on specifications, pricing, and timeline. It’s unclear whether Project Transformer is months or years from a potential launch, or whether it might ultimately be shelved like many internal Amazon projects. The company is known for its “working backwards” product development culture, where teams write mock press releases for products before building them. Not every mock press release becomes a real product.
Still, the strategic logic is hard to dismiss. The smartphone is the most important consumer device in the world. It’s the primary interface through which billions of people shop, communicate, consume media, and manage their lives. Amazon is the world’s largest online retailer and one of its largest cloud providers, yet it has zero presence on the device that mediates most digital commerce. Every purchase made through an iPhone enriches Apple’s services revenue. Every Google search on an Android phone feeds Google’s advertising machine. Amazon pays billions annually to be the default search result or app on other companies’ platforms. Owning the device itself — or at least having a meaningful foothold — would change that equation fundamentally.
There’s a cautionary note here, though. The AI capabilities that Project Transformer would depend on are still maturing rapidly. Alexa+, while more capable than its predecessor, has received mixed reviews since its launch. Some users have reported inconsistent performance, difficulty with complex multi-step tasks, and gaps in third-party integrations. Amazon has been iterating quickly, but the gap between a competent voice assistant on a smart speaker and a reliable AI agent on a smartphone is significant. On a speaker, a failed command is a minor annoyance. On a phone — the device people depend on for navigation, communication, payments, and emergencies — unreliability is unacceptable.
The competitive response would also be fierce. Apple and Google have spent more than a decade building mobile operating systems, developer relationships, and hardware supply chains. They have institutional knowledge about smartphone manufacturing, carrier relationships, and retail distribution that Amazon would need to either build or buy. And both companies are investing heavily in their own AI capabilities, which means any advantage Amazon might have with Alexa+ could be short-lived.
Then there’s the trust question. Amazon’s business model is built on knowing what consumers want to buy and making it frictionless to buy it. A smartphone that runs on Amazon’s AI would have unprecedented access to a user’s location, communications, browsing habits, and daily routines. Privacy-conscious consumers might balk at handing that data to a company whose primary business is selling them things. Amazon would need to address this head-on with transparent data practices and meaningful privacy controls — something the company hasn’t always excelled at with its Echo devices, which have faced scrutiny over recording practices and data retention.
For Amazon investors, Project Transformer represents both opportunity and risk. The opportunity is obvious: a successful AI smartphone could deepen customer engagement with Amazon’s retail, media, and cloud services in ways that no other product could. The risk is equally clear: another high-profile hardware failure would waste resources, distract management, and reinforce the narrative that Amazon can’t compete in consumer electronics beyond its narrow niches.
But Andy Jassy, who succeeded Bezos as CEO in 2021, has shown a willingness to make bold bets while simultaneously cutting projects that aren’t working. He’s wound down Amazon’s brick-and-mortar grocery experiments, scaled back the Scout delivery robot program, and refocused the company on AI and cloud computing. If Project Transformer doesn’t show internal promise, it likely won’t see the light of day. If it does, Amazon has the financial muscle — $73 billion in operating cash flow last year — to sustain a long and expensive market entry.
The smartphone industry has been waiting for its next major disruption. The shift from physical keyboards to touchscreens redefined mobile computing in 2007. The move to AI-first interfaces could be the next such inflection point. Whether Amazon — a company that failed spectacularly at smartphones the first time — is the one to lead that transition is an open question. But the fact that it’s trying again, with a fundamentally different approach and a fundamentally different technology stack, suggests that the lessons of the Fire Phone weren’t lost. They were just filed away, waiting for the right moment.
That moment, apparently, is now.
WebProNews is an iEntry Publication