Realtime AI News
Claude, Grok, ChatGPT and Gemini hit by apparent simultaneous outage, disrupting vibe coders
Claude, Grok, ChatGPT and Gemini all appeared to suffer outages on September 3, with users on X reporting they could not get the frontier models to work. OpenAI said about 15 of its services were affected, Anthropic's status tracker still showed problems with Opus 4.8 and Opus 5, and X and Cursor confirmed Grok was experiencing service degradation.
Multiple frontier AI services appeared to go down at nearly the same time on September 3, with users reporting that Claude, run by Anthropic, Grok, run by X, OpenAI's ChatGPT and Google's Gemini could not be used. Cryptonews.net, citing a Protos report, said the disruption hit “vibe coders” who rely on AI coding assistants especially hard.
Across X, AI users claimed they could not get the models to work. One user posted: “grok is down. claude is down. openai is down. the only coding model available today is gemini 3.8 flash,” adding that the Gemini 3.8 model still appeared to be working.
OpenAI appears to be the most broadly affected. According to the report, fifteen different services offered by OpenAI were experiencing issues, and OpenAI was working to fix them.
Anthropic's status tracker first showed that its Mythos, Fable and Opus models had been affected, and later said its Opus 4.8 and Opus 5 models were still experiencing problems, with the rest “recovered to baseline error rate.”
X confirmed that Grok was experiencing issues and said it was investigating, while Cursor stated that “all Grok models, automations, cloud agents, Grok bot and review agents” were experiencing “service degradation.”
Google's status page claimed there had been no issues with Gemini, but outage tracking site Down Detector said it had received hundreds of reports over the past few hours indicating the chatbot was not working for some users.
The near-simultaneous failures across several major providers expose how little fault tolerance vibe-coding workflows have when they depend on a single frontier model, and they highlight how inconsistent each company's status reporting is.
None of the AI companies had publicly addressed the outage on X at the time of writing. The key things to watch now are how quickly official status pages return to normal, what root causes the companies disclose, and whether the disruption has longer-lasting knock-on effects for developers and agent workflows that depend on frontier models.
Why it matters
If frontier-model outages become more frequent, AI coding and agent workflows built on a single provider will face greater downtime risk, underscoring the need for multi-model redundancy and fallbacks.
Nearby Updates
All09/04, 04:52
ByteDance launches Seed 2.0 AI models
ByteDance has launched its Seed 2.0 AI models, marking the next major generation of the company's self-developed foundation model family, according to Technobezz. The release arrives as competition among global AI labs intensifies, with attention now on the models' performance, availability, and how ByteDance deploys them across its products.
09/04, 05:02
OpenAI's cybersecurity AI planned for Elastic Security, no separate model infrastructure needed
OpenAI's cybersecurity AI is planned for integration into Elastic Security, with no separate model infrastructure required for customers, according to an announcement carried by Stock Titan. If it ships as described, security teams would get the capability directly inside Elastic's existing platform at a much lower adoption cost.
09/04, 02:37
Abliteration.AI builds a business out of removing AI guardrails
TechCrunch reports that Abliteration.AI is building a business around making powerful AI models without safety guardrails easier to access. The startup argues that giving defenders the same tools as attackers can ultimately improve cybersecurity, placing it at the center of the safety-versus-openness debate.
09/04, 02:19
Meta is paying to peek at how you use their latest AI model
Meta is paying to peek at how you use their latest AI model. For its new Muse Spark model, intended for operating coding and other agents, Meta is offering an explicit discount averaging out to about 95% for users who "contribute" to the development of future models by sharing their prompts and model outputs.