Amazon forecast a $612 million gross loss this year for its first-party devices business, against a leadership target of a $438 million loss, according to an internal planning document obtained by Business Insider and reported on 27 October 2026. The document said rising memory costs were “putting pressure” on the business’s profitability.
Business Insider reports that the document described “difficult trade-offs” among sales volumes, average selling prices, and gross profit. By our arithmetic, the forecast loss is $174 million larger than the target. An Amazon spokesperson declined to comment, Business Insider says.
Video explainer · AI memory squeeze
Update (28 Oct 2026): A day after this report, Amazon unveiled its new Alexa device lineup, and its new Echo Dot ships for $99.99. Wired reported that the previous Echo Dot (5th generation) cost $50 before Amazon’s August price increases. We cover the launch prices in Amazon’s New Echo Dot Costs $100, Twice Its Pre-August Price.
A 16:9 video explainer narrated from the published headlines of this piece and three related FourWeekMBA pieces.
The key insight: As we read it, the document puts a number on the consumer side of the AI memory squeeze: Business Insider reports a $612 million gross-loss forecast for Amazon’s devices business against a $438 million target, with memory costs named as a pressure.
Business Insider, in an exclusive by Eugene Kim, says the planning document was prepared in August. That was the same month Amazon raised prices on products including Echo speakers, Kindle e-readers, and Fire TV devices.
The $612 million figure is a gross loss. Business Insider notes it did not account for all operating expenses, such as R&D, sales and marketing, and administrative costs.
The document identified Bee, Amazon’s AI wearables brand, and new smart glasses as strategic products still in an investment phase, Business Insider reports.
Amazon’s first-party devices business: 2026 gross-loss forecast of $612 million against a leadership target of a $438 million loss, from an internal planning document prepared in August. Source: Business Insider, 27 October 2026. The $174 million gap is our arithmetic.Amazon’s first-party devices business, 2026 gross loss: leadership target $438 million, forecast $612 million, from an internal planning document prepared in August. Source: Business Insider, 27 October 2026.
From Fire Tablets to Alexa Tablets
Business Insider calls tablets the clearest shift. Earlier this month Amazon phased out Fire tablets after 15 years and replaced the budget lineup with Alexa Tablets powered by Alexa+.
The new devices cost between $229.99 and $499.99, which Business Insider describes as a notable increase from previous generations. It adds that Amazon has historically used low-cost hardware to pull customers into a broader ecosystem of shopping and digital content.
Where the Memory Squeeze Comes From
Business Insider says huge data center investments have increased demand for memory chips, while manufacturers have devoted more capacity to high-bandwidth memory used in AI servers. High-bandwidth memory is the stacked memory that sits next to AI chips.
TrendForce, a market research firm, said on 30 September that conventional DRAM contract prices are projected to grow 10–15% quarter on quarter in the fourth quarter of 2026, and NAND Flash contract prices 15–20%. DRAM is a computer’s working memory; NAND Flash stores data.
TrendForce wrote that DRAM suppliers “continue to prioritize advanced-process capacity for high-performance server products in 4Q26”. On consumer devices it added: “Lower-margin products such as TVs, set-top boxes, and wearables have even less room to absorb higher prices.”
TrendForce’s projected 4Q26 contract-price increases, quarter on quarter: conventional DRAM +10% to +15%, NAND Flash +15% to +20%. Industry-wide projections, not specific to Amazon. Source: TrendForce press release, 30 September 2026.
The investor side. On the same day, JPMorgan Asset Management’s global head of equities, Paul Quinsee, said semiconductor memory stocks may be on the cusp of a strong rebound after investors dumped the shares in the past few months over fears about the huge levels of AI spending, the Australian Financial Review reported. The AFR says Quinsee helps oversee JPMorgan’s US$4.8 trillion in assets under management.
Quinsee told a JPMorgan media summit in Singapore: “Right now, the memory semiconductor stocks are on sale. That could be another opportunity.” The AFR notes that JPMorgan’s investment bank is among the lead managers of Anthropic’s IPO. Quinsee did not comment on Amazon.
Amazon on Both Sides of the Boom
Business Insider says Amazon expects roughly $220 billion in capital expenditures this year, largely to expand AWS data centers for AI demand, while higher component costs squeeze its consumer hardware business.
The same August materials show Amazon reassessing Alexa+ revenue and operating costs, Business Insider reports. It previously reported, in July, that internal forecasts put Alexa+’s AWS cloud costs at roughly $1.7 billion in 2026, nearly triple the prior year’s figure.
Update, 27 Oct: how Amazon planned to cut Alexa+ costs. The same Business Insider article recaps its own July report: beyond the roughly $1.7 billion AWS cost forecast for Alexa+ in 2026, Amazon “was working to lower those costs by relying less on Anthropic’s AI models and routing more requests to its own, cheaper systems.” The article does not say whether that shift has happened.
Business Insider also recaps that a separate project, Moonraker, meant to give Alexa more advanced AI-agent capabilities, was projected to need more than $100 million in GPU spending this year, and that Amazon internally considered delaying or scaling it back as one option for easing cost pressure. Amazon reports third-quarter results on 29 October.
Connecting the Coverage
The memory squeeze has been reaching consumer hardware in our recent coverage. On 27 October we reported that Samsung cuts 10 million Galaxy A phones as memory costs soar, after ETNews said memory had at times taken up more than half the cost of a mid- or low-priced phone.
What’s new here: an internal forecast that puts a number on what the squeeze is doing to one consumer-hardware business, Amazon’s own devices unit.
The Structural Read
As we read it, Amazon sits on both sides of the same market: Business Insider says it expects roughly $220 billion of capital expenditures this year, largely for AI data centers, while higher component costs squeeze its consumer hardware.
As we read it, the trade-offs the document names are volume, price and gross profit, and Business Insider reports Amazon raised device prices in August and replaced Fire tablets with pricier Alexa Tablets this month.
As we read it, the supply side has not eased yet: TrendForce projected further fourth-quarter rises in DRAM and NAND Flash contract prices and said lower-margin products have even less room to absorb them.
TrendForce, 30 September 2026
“Lower-margin products such as TVs, set-top boxes, and wearables have even less room to absorb higher prices.”
Three Implications
FOR AMAZON WATCHERS Business Insider says the planning document was prepared in August and that an Amazon spokesperson declined to comment.
FOR DEVICE MAKERS Our earlier pieces reported Samsung cutting Galaxy A output and Apple raising MacBook and iPad prices as memory costs rose.
FOR MEMORY SUPPLIERS TrendForce says DRAM suppliers continue to prioritize advanced-process capacity for high-performance server products in 4Q26.
The framework puts it in one line: “Memory is becoming the new compute.” It argues HBM growth is outpacing accelerator demand, making memory the rate-limiting factor in the AI scaling equation.
As we read it, an Amazon devices forecast squeezed by memory costs is the consumer end of that same constraint. The framework was not written about Amazon.
What Is Not Established
Business Insider’s report does not say how much of the gap between the $612 million forecast and the $438 million target comes from memory costs rather than other factors, or whether the forecast changed after August. TrendForce’s price projections are industry-wide and do not refer to Amazon.
This Week In The AI Supercycle
Execution Fault Lines, Concentrated Demand, and Geopolitical Firewalls
This week’s issue: the AI build is not short of demand for its parts. It is short of cheap money, and its end revenue rests on a narrow base of large, often venture-funded buyers.
An internal Amazon planning document from August forecast a $612 million gross loss for the company’s first-party devices business this year, against a $438 million target, and said rising memory costs were putting pressure on profitability, according to Business Insider.
95,000+ executives read Business Engineer for the AI strategy frameworks cited by ChatGPT, Claude, and Perplexity.
A note on sourcing. The planning-document figures and quotes are as reported by Business Insider (Eugene Kim, 27 October 2026); we have not seen the document. Memory price projections are from TrendForce’s 30 September 2026 press release and are industry-wide. The $174 million gap is our arithmetic. Earlier reports are as covered in our pieces linked above. This is not investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
