Amazon Outage Map

Community Discussion

Tips? Frustrations? Share them here. Useful comments include a description of the problem, city and postal code.

Beware of "support numbers" or "recovery" accounts that might be posted below. Make sure to report and downvote those comments. Avoid posting your personal information.

Amazon Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Armaan Sidhu (@realarmaansidhu) reported 12 minutes ago

  Shay Boloor just made a specific forecast: Micron could be more profitable than Amazon AND Meta by 2027. That sounds absurd until you run the numbers. Here's what the memory cycle actually looks like right now. The receipts first. Micron's current annual net income (FY2025): roughly $8B. Amazon's current annual net income: roughly $62B. Meta's current annual net income: roughly $68B. Shay's claim: Micron's net income exceeds both by FY2027. For that to happen, Micron needs to roughly 9x its earnings in 24 months. Absurd? Micron grew earnings 284% in FY2025 alone. Sell-side consensus for FY2026 projects another 150-200% growth. Consensus for FY2027 projects the earnings acceleration continuing. Compound 200% YoY growth from $8B: FY2026 ~$24B. FY2027 ~$72B. The math works if the memory cycle continues. Why memory specifically is exploding. Every frontier AI model requires enormous amounts of memory bandwidth and capacity: Training: models need to hold billions of parameters plus gradient calculations in high-bandwidth memory during training runs. Inference: serving models at scale requires keeping parameters in fast memory for every query. The shift: GPUs can only be productive if memory can feed them data fast enough. Current high-bandwidth memory (HBM) pricing is up 300%+ in the last 18 months. Customers include Nvidia, AMD, Google, Amazon, Meta, OpenAI, Anthropic, Microsoft. Micron is one of only three companies globally producing HBM at scale (alongside SK Hynix and Samsung). HBM3e and HBM4 prices have moved from commodity to specialty pricing as demand exceeds supply. The demand forecast worth understanding. Morgan Stanley estimates global HBM demand will grow 3.5x between 2025 and 2028. SK Hynix (Micron's largest competitor) is sold out of HBM through 2027. Micron's HBM capacity is also effectively sold out through 2027 at current pricing. The pricing power this creates is what drives the earnings acceleration. When a company produces a commodity (traditional DRAM), margins are 10-20%. When a company produces a specialty product in structural shortage (HBM for AI), margins are 50-70%. Micron's margin structure is transitioning from commodity to specialty faster than most semiconductor cycles historically have. What could derail the thesis. Three specific risks: One: AI capex reduction. If Meta, Microsoft, Amazon, and Google collectively reduce AI infrastructure spending in 2026-2027, HBM demand contracts and pricing collapses. Current guidance suggests the opposite: Meta alone is spending $115-135B in 2026. But macro shocks could change this. Two: Chinese supply expansion. China's domestic HBM production (CXMT, Wuhan XMC) is scaling. If Chinese HBM reaches quality parity in 2027-2028, specialty pricing becomes commodity pricing faster than expected. Three: Capital return disappointment. Semiconductor companies historically convert earnings cycles into stock buybacks and dividends. If Micron's capital allocation disappoints, the multiple compresses even if earnings deliver. None of these risks are trivial. All are being monitored by institutional investors. Shay's specific claim requires HBM pricing to hold through 2027 and demand to grow consistent with current guidance. What this means for investors. Direct Micron exposure is one way to play this. Indirect exposure includes: SK Hynix (traded in Korea, ADR limited US availability). Samsung Electronics (conglomerate exposure dilutes memory thesis). Applied Materials, ASML, Lam Research (semiconductor equipment suppliers to HBM production). SMCI (Super Micro Computer — server manufacturing that uses HBM-equipped GPUs). DRAM ETF (DRAM): diversified exposure to the memory sector rather than Micron specifically. SMH ETF: broader semiconductor sector exposure including all HBM players. Which is the best play depends on conviction level and risk tolerance. The "Micron outearns Amazon and Meta" framing is a high-conviction single-stock bet. It can deliver spectacular returns if right. It can underperform dramatically if any of the three risks materializes. The DRAM or SMH ETF approach captures the structural HBM thesis without single-stock concentration risk. Most retail investors would be better served by the ETF approach, which we discussed earlier in the week regarding Intel's 26-year flat chart. Single-stock bets pay off spectacularly when they work and punish severely when they don't. The thesis underneath Shay's framing is right. The company capturing the most value from that thesis might not be Micron specifically. It might be SK Hynix. It might be Samsung. It might be a Chinese producer that reaches parity faster than expected. Sector exposure lets you win regardless of which specific company captures the value. Single-stock exposure requires you to pick correctly. For investors with genuine insight into HBM supply chains, Micron-specific exposure makes sense. For investors wanting to participate in the structural AI-memory thesis without claiming that specific insight, DRAM or SMH ETFs are the honest play. The memory cycle is real. The earnings acceleration is documented. The pricing power is structural. The question isn't whether it happens. It's how to capture it without concentrating risk in a single company whose execution you can't actually verify. Follow for more reads on where structural investment theses work but the specific implementation choices quietly determine whether you capture the return.

• 
  SparkyBru (@SparkyBru) reported 16 minutes ago

  @WallStreetApes @QVC has been financially in trouble for years…. Blame @amazon for businesses going into financial troubles…

• 
  BlueCollarMillionaire (@TheReal_BCM) reported 16 minutes ago

  @AmazonHelp Still waiting on you to issue a refund for wrong item shipped

• 
  GenZ Multi Family (@GenZMultifamily) reported 17 minutes ago

  @JulieChangRE nice. You should check out that rubber mat on amazon to prevent it from rusting. Mine rusted once and it was such a pain to fix.

• 
  Terrell_Mercer (@TMer25) reported 18 minutes ago

  @amazon why does your Prime subscription constantly go up and the quality of service goes down?

• 
  Jessica schmidt (@Jessica06562652) reported 19 minutes ago

  @apccustomz @WallStreetApes Why should I care they loose so much in shrink from loss .. It’s two ******* cents .. It’s literally a non issue. You pay more to use Amazon prime .. Again the world has changed and stupid *** people are stagnant. You think you get a full gallon of gas’s for the listed price … You don’t.

• 
  Nicholas Hyperion (@nickhistgeek) reported 21 minutes ago

  @DaveShapi Creative destruction? Try risk transfer. Amazon fires 30k while panicking over AI 'blast radius.' Entry-level down 14% since ChatGPT. Rational move: markets actively hiring like UAE.

• 
  Rod Beach (@Rodness1) reported 23 minutes ago

  @AmazonHelp It was still an issue. Didn’t matter what I did. Watching movies, no issue. Live broadcasts always an issue. And I have top-tier equipment. I have YouTube TV and watch live broadcasts in 4K and no issue. I watch UFC on Paramount+, no problem. It’s the same issues with TNF.

• 
  Sandy 〽️ (@RightSandy) reported 24 minutes ago

  .@MrsRotnjetski - You've been missing for a month from here! Praying everything is okay (and thank you for Amazon gift a couple of months ago - the gift tag was hidden under a box flap and I didn't see it until I broke down the box)

• 
  imsamit11 (@Imsamit11) reported 26 minutes ago

  @AmazonHelp It’s not a 'late shipment'—it’s a service deficiency. Your agent arrived but wasn't trained to handle the exchange code. Marking it 'Delivery Attempted' is a lie to cover the station's error. I expect this delivered today. Case ID: 3948bda5-3d8b-4335-8e9e-3d3961922a37

• 
  Huu Kh🅰️ng Investing (@Kan27111) reported 27 minutes ago

  @endless_frank What do you think the chance BO purposely fck up our stock, to delay our launching progress and give their Amazon Leo more time to playing catch up? It hard to think otherwise since they just buy GSAT last week. Or it is just a frustrating error in their launching business.

• 
  Mighty Joe (@joe1784) reported 28 minutes ago

  @flipkartsupport 2 days I have waited for the product and not stepped out for any work in order to avoid any delivery delay or issue but in return I got nothing only waste of my valuable time and no product in hand. No wonder Amazon is ahead of you . What a shame!! @Flipkart @jagograhakjago

• 
  Leslie (@vallery78955) reported 30 minutes ago

  @Too_Much_Rum I personally want my package delivered to my protected front porch or carport, especially if it's raining. I have no problem with delivery because Amazon built a new distribution center less than 2 miles from my house so some things are delivered same day! I love it. No drones.

• 
  TerminatorMom (@MomTerminator) reported 31 minutes ago

  @DanMiressiCT I'd be happy if we had stores again! With products on the shelves. Lamont shut down our state and we lost the malls and mom and pops. We are left with Walmarts and Amazon.

• 
  Jose M (@JMLV51) reported 36 minutes ago

  Amazon saved about $4B in taxes and still cut 30,000 jobs. Verizon saved $2B and plans 15,000 layoffs. Meta saved $3B and may cut 20%+ of its workforce. Meanwhile I paid more in taxes + higher gas, groceries, utilities, and car insurance. And we’re told trickle down works… for who exactly?