OpenAI CEO Sam Altman speaks in the course of the OpenAI DevDay occasion on November 06, 2023 in San Francisco, California.
Justin Sullivan | Getty Pictures
ChatGPT, OpenAI’s viral chatbot, is experiencing outages attributable to a focused assault, in accordance with an OpenAI spokesperson. As of Thursday, customers have been nonetheless receiving some error messages.
Greater than 92% of Fortune 500 corporations use the platform, up from 80% in August, and so they span throughout industries reminiscent of monetary companies, authorized purposes and training, OpenAI Chief Expertise Officer Mira Murati advised reporters Monday.
“We’re coping with periodic outages resulting from an irregular visitors sample reflective of a DDoS assault,” OpenAI shared on Wednesday night, simply earlier than 11 p.m. ET. “We’re persevering with work to mitigate this.”
The corporate mentioned buyer info had not been compromised.
The problems comply with Wednesday’s outage throughout the corporate’s instruments and companies, together with the API utilized by greater than two million builders, and appeared to start Wednesday morning, when OpenAI reported errors “impacting all companies,” in accordance with its standing web page. ChatGPT customers have been advised “ChatGPT is at capability proper now,” however after implementing a repair, issues returned to regular, per OpenAI — however not for lengthy.
The assault additionally follows OpenAI’s first in-person occasion on Monday, the place the firm introduced it had surpassed 100 million weekly lively customers. It additionally introduced its strongest synthetic intelligence mannequin but, GPT-4 Turbo, and a brand new choice permitting customers to create custom-made variations of ChatGPT.
