OpenAI says abnormal traffic is causing outages for its ChatGPT, API, Labs and Playground services and that it is trying to mitigate the ongoing issue.
OpenAI claims to be suffering from a distributed denial-of-service (DDoS) that is causing periodic outages for the company and its products.
The company began investigating the issue at roughly 8pm GMT yesterday (8 November) and claimed it implemented a fix within one hour, according to an incident report.
But by roughly 1am GMT OpenAI said it was still seeing “periodic outages” across ChatGPT and its API. The company said these outages are also impacting its Labs and Playground services.
“We are dealing with periodic outages due to an abnormal traffic pattern reflective of a DDoS attack,” OpenAI said in the latest update on the report. “We are continuing work to mitigate this.”
A DDoS attack is an attempt to make an online service unavailable by overwhelming it with high volumes of data from multiple sources. Typically, multiple compromised computer systems are used as sources of attack traffic.
OpenAI has not disclosed the scale of the suspected attack. A member of its developer relations team, Logan Kilpatrick, said the “abnormal traffic” being experienced is reflective of a DDoS attack.
Jake Moore, a global cybersecurity advisor with software company ESET, said DDoS attacks are a clever way of targeting a company without having to hack the mainframe and that the perpetrators can remain “largely anonymous”.
“This makes it that much more difficult to protect from when the landscape is completely unknown apart from having the strongest DDoS protection available,” Moore said. “Unfortunately OpenAI remains one of the most talked about and current technology companies making it a typical target for hackers wanting their own kudos.”
The suspected attack follows OpenAI’s first-ever developer conference earlier this week, where it claimed ChatGPT has reached the milestone of 100m weekly users. The company also announced a new feature that lets users build their own custom ChatGPT models for various purposes.
In September, the Silicon Valley start-up announced it is opening a new office and creating nine new roles in Dublin as it eyes further expansion in Europe.
But new competition is also on the horizon for the company, as Elon Musk’s start-up xAI recently shared details of its own chatbot called Grok that is being advertised as a chatbot with a “rebellious streak”.
The year of DDoS
As technology becomes more advanced, the power of cyberattacks also seems to be on the rise. This appears true for DDoS attacks too, as last month multiple Big Tech firms claimed they mitigated record-breaking DDoS attacks.
For example, Google said it mitigated a DDoS attack that generated 398m requests per second earlier this year, while last year’s record was only 46m requests.
A report by Netscout claimed there were 7.9m DDoS attacks in the first half of 2023, representing a 31pc increase compared to the same period in 2022. The report claimed the rise in attacks was linked to global issues, such as Russia’s invasion of Ukraine.
10 things you need to know direct to your inbox every weekday. Sign up for the Daily Brief, Silicon Republic’s digest of essential sci-tech news.
