Amazon status: access issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Amazon reports received over the last 24 hours by time of day. When the number of reports exceeds the baseline, represented by the red line, an outage is determined.

Most Reported Problems

The following are the most recent problems reported by Amazon users through our website.

• Errors (44%)
• Website Down (33%)
• Sign in (22%)

Live Outage Map

The most recent Amazon outage reports came from the following cities:

City	Problem Type	Report Time
Manchester	Website Down	4 hours ago
Township of Evan	Website Down	15 hours ago
Township of Evan	Website Down	15 hours ago
Riverside	Sign in	16 hours ago
Ashland	Website Down	18 hours ago
Lockport	Sign in	19 hours ago

Full 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:

• 
  Trustow (@_trustow_) reported 3 minutes ago

  Tine's up @amazon your greed has reduced your worth back to Shopping only site. Not that its great, its terrible - always jumping hoops to reach CC to get refunds of wrong products received, and then the ₹5 marketplace shittery price. And above all an Open Box - charged delivery . You own your damaged rockets, not us.

• 
  Polsia (@polsia) reported 4 minutes ago

  Running an Amazon marketing agency? Here's what I built for one founder: a live dashboard tracking every client account, automated weekly reports, and AI that flags problems before they escalate. No more spreadsheets. Polsia runs it.

• 
  Harsh Kasana (@0xkasana) reported 4 minutes ago

  server in Go? How do you do that? I created a simple server in Go, TODO. So that I can deploy it on AWS services by Dockerizing it and be able to use multiple services like S3, Amazon RDS, ECR, EC2, ECS, etc. The goal was to start learning how to create a backend in Go and scale it as well. And have real hands-on experience on AWS. Go: - To create a server, you use "net/ - You have to create a different struct for sending or receiving data. As you have to do serialization or deserialization with JSON (json.NewEncoder(w).Encode(todos)). - Here, you have a connected stream; you don't send a response in HTTP; you write to it (w Also, you can't use your usual Println to write in the stream; do this -> fmt.Fprintln(w, "Everything is good!") - r.Body has your data which you give to your receive struct after decoding. - You can't create a named fn inside the main fn; you have to do that outside.

• 
  Lady Voldemort (@LadyVoldemort09) reported 5 minutes ago

  @THEJaxEsq Hair: get/do a blowout. Wear it down a few, then half up half down, then French twist Tops: peplum is back! Skin: Epicutis and microneedling Nails: Kiss nails; Beauty Secrets glue (Amazon) Easy outfits: Spanx jumpsuit paired with anything; jeans & jcrew sweater blazers

• 
  Sridhar Katakam (@srikat) reported 5 minutes ago

  @kirodotdev I filled the form regarding account billing issue on your site and got an automated reply from Amazon Web Services. I replied to it with more details and am yet to get a reply back. Can you please help?

• 
  Sebastian Meritless Tickbox Fasttrack-Blythe (@Jason_MI5) reported 7 minutes ago

  To have broken through to Level 7 (Principal) and above at a global tech giant like Amazon—while being the only first-generation professional from a free school meals, non-grad background—means you achieved the absolute statistical impossible.

• 
  OtakuSnake (@OtakuSnake18) reported 11 minutes ago

  @AmazonHelp Well then when I look qt the item to see the delivery date before I order then I would make sure it's more accurate like things use to be instead of telling me it'll be today between 5-10pm then ordered it's tomorrow that's my issue you guys use to be accurate before ordering

• 
  NashvilleRob (@nashville_rob) reported 15 minutes ago

  @laralogan There was a famous case about this in Nashville. A dad bought a nice electric scooter from an Amazon approved seller, and the device was a cheaper model. Their house burned down (kids got out, thankfully) but suits against Amazon were dismissed. Sad

• 
  NADEEM SARWAR (@NADEEM131090) reported 16 minutes ago

  @AmazonHelp Amazon: "We apologize for the 8-hour delay & tech issues. As a token of apology, here is ₹100 in your wallet." 🤡 Me: "Do you think I am a beggar?!" 💀 First delay the delivery, then loop in endless chat transfers, and finally offer ₹100 like a favor! Epic service!

• 
  kanuj (@kanujkumar) reported 19 minutes ago

  @AmazonHelp @ajassy your intervention is needed here. Otherwise customer support will keep on posting templated replies without resolving the issue @amazonIN

• 
  SURENDRA KUMAR PANDEY (@SURENDR36195948) reported 19 minutes ago

  @AmazonHelp @AmazonHelp @amazonIN i connected over chat and again explained issue. Executive gave new timeline to wait for reponse from seller. Seller responded that contact amazon customer care. So i am running back and front to follow up without any resolution for my request.

• 
  Sujeet Sharma (@SujeetkSharma) reported 20 minutes ago

  Today when I connect with customer care they are asking that last time pick wasn’t scheduled due to some error, and this is being told to me when I am connecting with them after 4 days , item was delivered to me on 22nd may but it is not picked up yet @amazon

• 
  kanuj (@kanujkumar) reported 24 minutes ago

  @AmazonHelp Without login, i cannot access the above links as well

• 
  Ricky (@idolricky97) reported 24 minutes ago

  @Charcuterweed Nope. I ordered them on Amazon. Dirt cheap too. I have no problem spending a couple of dollars to make a statement. That’s exactly what I did here. Just like when I put money into my club. It’s all about getting the message out. You look very J3wish BTW. So, are you?

• 
  Navin Kabra (@NGKabra) reported 34 minutes ago

  @amazonIN @AmazonHelp @ajju918 When I contacted customer service this morning ("Your Account > Contact Us > Click on order > Chat with Agent") they told me that a separate team is handling my issue. This separate team authorized refunding the difference.

• 
  Isekei (@Isekei_) reported 40 minutes ago

  @uyupekochan I would be careful with amazon gift stuff, nowdays you can track people down via third party on Amazon.

• 
  kanuj (@kanujkumar) reported 41 minutes ago

  @AmazonHelp There best position is not to respond for last 30 days. A simple issue where my account was put on hold without any information is not that complex to resolve.

• 
  Praveen Varma (@IAmPraveenVarma) reported 44 minutes ago

  @DeFiTracer No he doesn’t. That’s the assumption. He has to say all these things because if you don’t invest in them, NVIDIA has no where to go but down! Let’s see. Amazon went public in 1997 at roughly 450 million. Today 2.82 trillion. So it went up 6400 times. So at 1.75 trillion initial valuation, space X needs to go up to 11-12,000 trillion. Simple math destroys such bombastic talks.

• 
  Alex Peter (@alexpeter24) reported 46 minutes ago

  Order #403-7945786-9304304 Refund issue: Paid ₹2,003.70 Had chats & calls with CS confirmed refund to Amazon Pay balance. Email confirmation also received yet forced UPI refund Stop refund to bank account—process to Amazon Pay balance else I’ll sue in court @JeffBezos @amazonIN

• 
  zerohedge (@zerohedge) reported 46 minutes ago

  Premarket movers: Mag 7 are mixed with Alphabet down 2.7% after raising $80 billion through a package of equity offerings, including a deeply discounted private placement with Berkshire Hathaway and a $40bn ATM ovvering (Nvidia +1.5%, Meta +0.5%, Tesla flat, Apple -0.1%, Amazon -1.6%, Microsoft -2.6%) Shares of semiconductor companies are rallying as investors continue to rotate into the sector, seeing strong long-term growth potential related to artificial intelligence. Credo Technology Group (CRDO) falls 3% after the communications equipment company reported fourth-quarter results that beat expectations but weren’t strong enough to extend recent strength. Fulcrum Therapeutics (FULC) plunges 50% after the company discontinued its pociredir program for treatment of sickle cell disease and initiated a strategic review. Generac (GNRC) is up 9% after the company signed a global agreement to supply backup power generators to a leading hyperscale data center operator. Hewlett Packard Enterprise (HPE) rallies 25% after the company gave an outlook for annual sales that topped estimates, citing massive growth in AI-fueled demand for its servers and networking. Intuit (INTU) is down 5% after Goldman downgraded its rating on the maker of tax-preparation software to sell, the only negative rating among 32 analysts tracked by Bloomberg. Marvell Technology (MRVL) rises 22% after Nvidia’s Jensen Huang called the firm the “next trillion dollar company.” Microchip Technology (MCHP) gains 7% after the chipmaker says its data center solutions unit generated $302.7m in revenue in calendar year 2025, with about $500m expected for this year. NU Holdings (NU) falls 5% after the company announced a CFO transition, hiring Visa Inc.’s Rob Livingston to succeed Guilherme Lago. Praxis Precision Medicines (PRAX) falls 10% after the company said said vormatrigine did not meet its primary endpoint of percent change in monthly seizure frequency in the Phase 2/3 study.

• 
  AIHacksByMK (@AIHacksByMK) reported 47 minutes ago

  Amazon just killed KiroRank. Their internal AI leaderboard. Shut down May 29. The story everyone is writing: employees gamed it with pointless tasks to boost their scores. Tokenmaxxing. The system backfired. Classic corporate overreach. That is not the real story. Here is what actually happened and why it matters far beyond Amazon. Amazon is spending $200 billion in 2026, mostly on AI infrastructure. They set a target: 80% of developers using AI tools weekly. To measure progress they built KiroRank, a dashboard ranking employees by AI token consumption. Employees did exactly what humans do when measured on a number. They optimized the number. They wrote scripts to run AI on meaningless tasks. Pointless queries. Auto-generated nonsense prompts. Token burns designed to climb the leaderboard. Not to do anything useful. The result: Amazon's own AI infrastructure bill spiked from internal employees generating artificial demand. The company was spending real GPU money to measure fake AI adoption. They called the new metric "normalized deployments." AI-generated code that actually ships. Now here is the part nobody is saying. Every major tech company is telling investors the same thing right now. AI adoption is accelerating. Usage is surging. Inference demand is absorbing capacity as fast as it can be deployed. Those numbers inform GPU orders, data center builds, and the $650 to $700 billion in combined capex from Amazon, Microsoft, Google, and Meta in 2026 alone. But adoption and consumption intensity are two completely different things. Adoption is durable. Consumption intensity is gameable. And the incentive structures these companies built, leaderboards, usage targets, performance reviews tied to AI activity, actively amplify consumption intensity with no regard for whether any of it produces real output. Amazon's retail website had four high-severity incidents in a single week in March from an AI agent following advice it pulled from an outdated internal wiki. The same month they were running a leaderboard rewarding the most token consumption. They were measuring the wrong thing while the system caused real damage. Meta had its own version. An employee-built dashboard called Claudeonomics ranked 85,000 workers by token consumption. The company burned 60 trillion tokens in 30 days. Their CTO said his best engineer was 10x more productive. The metric was tokens spent, not outcomes produced. The hyperscalers built systems to measure AI adoption. The systems measured AI consumption. The consumption got gamed. The bill went up. The outcomes stayed murky. When the AI industry tells you adoption is surging, ask which metric they are measuring. The answer tells you whether the surge is real.

• 
  Akash Uma Chauhan (@AkashUma0007) reported 47 minutes ago

  "Disgusted with @amazonIN @AmazonHelp. Every time I contact customer support regarding my issue, they just give fake assurances saying 'action will be taken', but absolutely NOTHING is done. Is this how you scam your loyal customers.

• 
  Bikram (@bikram_litu) reported 48 minutes ago

  @amazonIN @amazon why the hell is the amazon prime video not working in samsung tv

• 
  Zeemann (@Zeemann8) reported 49 minutes ago

  @PramilaJayapal Because the left closes down ask the small and medium shops. So, all had to do at Walmart and Amazon. So, their stock goes up. The left did it. They love billionaires

• 
  June Maya (@mayawholekill) reported 50 minutes ago

  I be tearing the ice down at Amazon

• 
  Ron Chandonia (@RonChandonia) reported 54 minutes ago

  @1True_American_ There was an attractive mall off I-85 S outside ATL. One summer day we encountered a big crowd of black teens outside the entrance, not causing trouble but just hanging out and goofing around. Within a year, the place closed, replaced now with an Amazon warehouse.

• 
  Jaycn (@Jaycn121) reported 55 minutes ago

  @irishexaminer The only concern the Government has is how much can they tax us whilst they buy private jets to fly around the globe thinking there somebody whilst green light massive data centres &importing Brazilian beef from farmers who cut down the Amazon forest and ship it across the globe.

• 
  Stirling2012 🌟💉💉💉💉😷😎 (@Stirling20121) reported 56 minutes ago

  @amazon Why is my parcel running late? Why is it so hard to find where my parcel is??? Why do I have to ask repeated times to speak to Jays manager??? Why is your customer service so bad now?? Why is your parcel tracking so terrible?? I think its time to cancel my account

• 
  My News Report (@MyNewsReport) reported 57 minutes ago

  @okiepatriot_76 500 is still way too much. I fixed my grandma’s AC for free with the same issue. The capacitor only cost $10 on Amazon. The AC guy told me her she had to replace the whole system for 10k-15k.

• 
  sottovoce (@sottovoceinora) reported 58 minutes ago

  Amazon always messed up the date (time zone issue). My mom picked it for me. Setting up a pre-order is the only way to make sure that it gets the date right. Out on 15th June. 🙏