OpenAI posts dual major incidents; Sora failures coincide with widespread request timeouts, pressuring integrations
Published Mar 18, 2026, 6:31 PM UTC
Key entities
TLDR
Treat OpenAI as degraded across modalities: keep non-Sora workloads on fallbacks and circuit breakers, and assume partner integrations like Walmart’s are sensitive to renewed instability until OpenAI provides root-cause and recovery details.
Why this matters
Inference: The overlap of a broad timeout incident and a modality-specific Sora failure suggests platform-level reliability risk rather than an isolated Sora issue (medium confidence). Rationale: One incident affects general request handling, while another targets Sora video; concurrent major-severity postings increas…
What changed
- OpenAI reports two major incidents in close succession: user requests timing out/content not loading (identified) and Sora video generation failing (investigating).
- Wired highlights Walmart’s shift to embed its Sparky shopping agent directly into ChatGPT and Google Gemini after OpenAI’s Instant Checkout fell short.
- Incident 1: “User Requests Timing Out / Content Not Loading” marked Major, status Identified.
- Incident 2: “Sora video generation failing” marked Major, status Investigating.
Topic context
Use this page when you want fast context on confirmed model launches from OpenAI, Anthropic, Google DeepMind, xAI, Meta, and similar labs without scanning every release note, model card, or developer post yourself. Key angles: openai, anthropic, google deepmind, gemini.
Summary
OpenAI simultaneously lists a major Sora video generation failure and a separate major incident for user request timeouts/content loading, suggesting risk beyond a single model path; while Wired reports Walmart is embedding its Sparky shopping agent into ChatGPT and Google Gemini after prior feature shortfalls, the current instability raises near-term availability risk for such integrations until OpenAI clarifies scope, cause, and mitigation timelines.