Emerging Threats

OpenAI Outage Disrupts ChatGPT Ahead of Astra Model Launch

Analyst 207||Source: BleepingComputer
Empty tech office with error messages on screens and natural daylight pouring in.

"We are investigating the issue for the listed services," OpenAI said.

OpenAI acknowledgment and ongoing investigation

OpenAI confirmed a major outage affecting ChatGPT and Codex on its status page and said it is investigating. The company acknowledged elevated errors across multiple services and components; as of the time of writing, the investigation was still underway. OpenAI's status update is the clearest public acknowledgment available while engineers work to restore service.

When the outage began and how broad it is

The interruption began at approximately 10:58 AM ET on Thursday, September 3. Reports and OpenAI's status information indicate the problem is not isolated: it spans "nearly every major ChatGPT feature" and includes OpenAI's Codex services. According to the status page, at least 15 ChatGPT components are affected.

Services and features reported down

The status listing names affected components explicitly: Conversations, Login, ChatGPT Work, Codex in ChatGPT Desktop, Compliance API, Search, File uploads, Voice mode, GPTs, Image Generation, Deep Research, Agent, ChatGPT Atlas, Sites, and Connectors/Apps. Users have reported a range of failures: ChatGPT may fail to load conversations, return errors when sending messages, or prevent use of capabilities such as Search, file uploads, image generation, Deep Research, and Voice mode.

Astra model launch is imminent but no link confirmed

The outage occurs ahead of the expected launch of OpenAI's next major AI model, Astra, which the company recently confirmed will be released soon. OpenAI has not said whether the September 3 outage is related to preparations for Astra, and the company or reporting notes that it is unlikely the issues are related, "as ChatGPT has frequent outages." That distinction leaves the timing notable but unproven as causally connected.

What this means for end users, enterprises, and technologists

  • End users: Individuals relying on ChatGPT for conversations, image generation, voice interactions or file uploads may see conversations fail to load or receive error messages. The outage impairs a broad set of interactive features.
  • Enterprises and procurement leaders: Organizations using ChatGPT Work, ChatGPT Atlas, Connectors/Apps, or the Compliance API should expect interruptions to workflows tied to those components until OpenAI resolves the incident.
  • Technologists and security teams: Engineers and platform operators observing downstream failures should note that Codex services and desktop integrations are affected, and that troubleshooting will be constrained until OpenAI's remediation completes and the company provides further updates.

OpenAI's public status page remains the primary source for progress updates; the company has acknowledged the outage there and is "investigating the issue for the listed services." Whether the elevated errors will delay or otherwise affect the Astra launch is not addressed in OpenAI's statements and remains an open question.

For the time being, users and organizations encountering errors will have to wait for OpenAI's investigation to conclude and for service restorations to be posted to the status page. The close temporal proximity to the announced Astra launch raises questions about timing, but the company has not connected the two publicly and the available information describes the outage as one in a pattern of frequent ChatGPT interruptions rather than a confirmed pre-launch event.

Original story: https://www.bleepingcomputer.com/news/artificial-intelligence/openai-confirms-chatgpt-is-down-ahead-of-astra-model-launch/

Artificial IntelligenceChatgptEmerging ThreatsOpenaiService OutageCodexApplication Downtime
Related Intelligence

Continue Reading

Technicians walk through a brightly-lit server room with error messages on screens and a person's hand reaching toward a…

Anthropic Outage Disrupts Multiple AI Models

Big trouble hit Anthropic on September 3, 2026, when a widespread outage caused multiple AI models, including Claude Mythos 5.1, Claude Fable 5.1, and Claude Opus 5, to produce elevated errors, disrupting users at 9:41 AM ET. The company quickly identified the issue and was working on a fix, but the outage remained ongoing.

Analyst 207
Young adult sits in quiet setting, surrounded by books and natural light, with subtle hint of technology.

Pegasus Spyware Targets Serbian Student Activist via iMessage Exploit

A shocking spyware attack has been uncovered, where a Serbian student activist was secretly infected with Pegasus spyware through a clever iMessage exploit that required zero clicks from the target. This sinister attack allowed the spyware to be delivered undetected, putting the activist's personal data and safety at risk.

Analyst 207
Server cabinet with exposed internal components in a small business server room.

Elementor Pro Exploit Targets WordPress Sites with Webshell Payload

A critical flaw in Elementor Pro, tracked as CVE-2026-32475, has put thousands of WordPress sites at risk, with nearly 200,000 attempted exploits blocked by Wordfence's web application firewall since August 19. This vulnerability allows attackers to bypass file-upload validation and inject a malicious PHP payload, compromising site security.

Analyst 207
Rows of rack-mounted servers in a dimly lit, cramped server room.

Outsider Phishing Kit Persists After Takedown Effort

Despite a major takedown effort that seized servers, domains, and $100,000 in payment wallets, the Outsider Phishing Kit is still up and running, with over 700 new phishing pages created within just a month. This alarming resurgence shows that threat actors are relentless in their pursuit of online deception.

Analyst 207