Reading mode:
Key Takeaways
  • Apple projects its gross margin to decline to 45-46% in the holiday quarter, down from 47.9%, due to increasing memory costs and Siri AI expenses.
  • The company reported record September-quarter revenue of $114.5 billion, up 12%, with Mac sales surging 24% year-over-year.
  • CEO John Ternus described current memory price increases as “unprecedented,” with manufacturing purchase obligations climbing by $33 billion.

Apple Forecasts Margin Decline Due to Surging Memory Prices and AI Investments

Apple has cautioned that its gross margin is expected to fall to between 45% and 46% in the upcoming holiday quarter, a decrease from 47.9% reported in the September quarter. This anticipated contraction is primarily attributed to the unprecedented rise in memory chip prices and increased expenditures associated with the development of artificial intelligence for Siri.

The announcement was made during the company’s quarterly earnings call, following the release of its record-breaking financial results for the fiscal fourth quarter of 2026. For more details on Apple’s overall financial performance, see our report: Apple Reports Record Q4 2026 Earnings: $114.5 Billion Revenue, Up 12%.

Unprecedented Memory Cost Increases

Apple CEO John Ternus characterized the current surge in memory pricing as “unprecedented.” While the company managed to partially offset the impact in the September quarter through existing inventory, lower costs on non-memory components, and price adjustments on some products (Macs and iPads in June, iPhones in September), even higher memory costs are expected in the December quarter. CFO Kevan Parekh indicated that memory costs alone led to an approximate one percentage point drop in gross margin between the June and September quarters.

To secure supplies, Apple’s manufacturing purchase obligations jumped by $33 billion between June and September. Memory suppliers such as Samsung, SK Hynix, and Micron have also warned that supply shortages and price inflation, driven by robust demand from AI data centers, could extend into 2027 and 2028.

Siri AI Expenses Weighing on Profitability

In addition to memory costs, the expansion of Siri AI capabilities is expected to exert further pressure on margins. While specific expenditure details were not fully disclosed, Apple is actively enhancing its AI offerings, positioning the Mac as a powerful platform for running local AI models. Ternus highlighted Mac mini and Mac Studio as devices pushing the boundaries of local private AI.

Booming Mac Business Amid Supply Constraints

Despite these challenges, Apple’s Mac business is thriving. The company reported selling more Macs in fiscal 2026 than in any previous year. In the September quarter, Mac revenue reached $10.8 billion, marking a 24% year-over-year increase. However, COO Sabih Khan confirmed that Apple experienced supply constraints across iPhone, Mac, and iPad in the September quarter due to limited advanced chipmaking capacity. For the December quarter, constraints are specifically anticipated for Mac and iPad, primarily driven by memory shortages due to higher-than-expected demand.

Metric September Quarter (2026) December Quarter (2026) Forecast
Total Revenue $114.5 billion (+12% Y/Y) 7-10% Growth
Mac Revenue $10.8 billion (+24% Y/Y) Supply Constraints Expected
iPhone Revenue $57.6 billion (+17% Y/Y) -
Services Revenue $31.6 billion (+10% Y/Y) -
Gross Margin 47.9% 45-46%
Manufacturing Purchase Obligations Increase (June-September) +$33 billion -
What This Means in Practice

For consumers, the combination of rising memory costs and AI development expenses from Apple could translate into further hardware price increases or a slower rollout of new features. For investors, it signals potential pressure on profitability despite overall revenue growth. The broader industry faces sustained demand for memory from the AI sector, leading to ongoing shortages and component price inflation for all electronics manufacturers.

Interactive Hardware & AI BenchmarkFull VRAM & ROI Calculator →
Can Your GPU / AI Station Run 2026 Local AI Models?
✓ Optimal (100% GPU VRAM, ~55-85 tok/s)ollama run qwen3:14b