A widespread Amazon Web Services (AWS) outage, already covered in detail by TechJuice[1], left millions of users offline on Monday, affecting major platforms like Snapchat, Reddit, Duolingo, Fortnite, Airbnb, and Disney+.
Amazon confirmed that the disruption stemmed from a Domain Name System (DNS) issue. Engineers managed to resolve it within hours, but AWS said there was still a “backlog of requests” being processed as services recovered.
This outage once again highlighted how dependent global platforms are on AWS cloud infrastructure. Even a short disruption rippled across businesses, entertainment platforms, and messaging services worldwide.
Pakistan Faces Double Trouble with Internet Outage
In Pakistan, the impact was far worse. Alongside the AWS issue, users across Karachi, Lahore, and Islamabad faced slow speeds and connectivity drops. According to several Internet Service Providers (ISPs), this was due to a fault in the PEACE submarine cable system, which connects Pakistan and China.
Digital expert Habibullah Khan shared on X:
“It’s a cable cut on the core system. They are running tests to localise the fault location.”
Internet bad? The PEACE cable is down. It’s a cable cut on the core system. They are running tests to localize the fault location. Will update you once the location is confirmed. pic.twitter.com/qgdJghW42c[2]
— Habibullah Khan (@Huk06) October 20, 2025[3]
Despite user complaints, there was no official confirmation from local ISPs, the Pakistan Telecommunication Authority (PTA), or the IT Ministry.
When approached by reporters, Minister for IT and Telecom Shaza Fatima Khawaja said:
“Snapchat and other major apps [are] down due to Amazon Web Service outage worldwide.”
According to the PTA, submarine cable operations in Pakistan are handled by PTCL, CyberNet, and Transworld Associates. PTCL operates three international cables, AAE-1, SMW4, and IMEWE, that carry most of Pakistan’s internet traffic. Any fault in these systems can lead to nationwide slowdowns.
ChatGPT Users Also Report Outages
While AWS engineers were troubleshooting, another major platform began showing problems: ChatGPT.
Users from Asia, Europe, and North America flooded Downdetector with outage reports, saying the chatbot wasn’t responding or was showing “Error in message stream” warnings.
OpenAI has yet to issue an official statement. However, since ChatGPT relies heavily on AWS cloud infrastructure, experts believe the outage may have indirectly affected AI-based services as well.
In Pakistan, several users reported that ChatGPT and related OpenAI tools were inaccessible or running slowly during the same time window, suggesting a combined impact of AWS disruption and the local cable fault.
A Global Internet Wake-Up Call
Monday’s incidents serve as a reminder of the world’s growing reliance on cloud-based infrastructure. A single outage can paralyze communication, AI services, and digital transactions globally.
In Pakistan, where connectivity depends on a few undersea cables, even minor faults can trigger major slowdowns. Experts urge the government and ISPs to diversify bandwidth sources and improve transparency during outages.
There was no statement or acknowledgement from the ISPs in question, the IT ministry, or the regulator, Pakistan Telecommunications Authority.
As AWS, OpenAI, and local telecom operators restore full services, users continue to call for better redundancy in both global cloud networks and Pakistan’s submarine cable systems.
References
- ^ TechJuice (www.techjuice.pk)
- ^ pic.twitter.com/qgdJghW42c (t.co)
- ^ October 20, 2025 (twitter.com)