Anthropic confirmed widespread service disruptions across multiple Claude AI services on Tuesday, causing significant access issues for thousands of users. According to Anthropic's official status page, the company acknowledged the outage stating "The issue has been identified and a fix is being implemented. We are currently investigating this issue." The disruption affected Claude AI, Claude Chat, Claude API, Claude Console, and Claude Code services, with users encountering failed requests, slow loading times, and intermittent service interruptions. As reported by Downdetector, Claude AI experienced 279+ reports of outage in India as of 1:13 pm, with Claude Chat accounting for 61% of complaints, app issues affecting 20% of users, and website problems reported by 14% of users. Users first began reporting issues on Tuesday morning, with many saying Claude AI was either not responding or displaying error messages, while some were unable to log in or generate responses.
Anthropic officially released Claude Opus 4.8 on May 28, 2026, just weeks after the April launch of Opus 4.7. According to reports from Anthropic, the release confirms earlier speculation following leaks on reddit that suggested a planned rollout. The new model is now available at the same price as its predecessor with powerful new features for coding, agentic workflows, and user control. The latest service disruptions occurred on June 1, 2026, with multiple incidents affecting different Claude models throughout the day. As the outage continued, users flocked to social media platforms to check whether others were facing similar issues, with many sharing screenshots showing failed prompts, loading screens and error notifications.
Opus 4.8 demonstrates stronger performance across coding, agentic skills, reasoning, and practical knowledge work benchmarks. As reported by Anthropic, early testers highlight greater reliability, sharper judgment, and significantly improved honesty. The model is four times less likely than Opus 4.7 to miss flaws in code it produces and is less prone to unsupported claims. With adaptive thinking enabled, Claude Opus 4.8 triggers reasoning only when it judges the turn needs it. On simple lookups and short agentic steps it responds directly; on complex multi-step problems it reasons before answering, reducing wasted thinking tokens on bimodal workloads compared to Claude Opus 4.7 at the same effort level.
On the Legal Agent Benchmark, Claude Opus 4.8 delivers the highest score recorded and is the first model to break 10% overall on the all-pass standard. According to Anthropic, the model carries end-to-end tasks through with greater efficiency and tool calling using fewer steps for the same intelligence. For substantive legal work, this accuracy lift translates directly into how much real attorney work customers can hand off with confidence. The model also scores 84% on Online-Mind2Web, representing a meaningful jump over both Opus 4.7 and GPT-5.5, making it the strongest computer-use and browser-agent model tested.
The latest release introduces several significant improvements including Effort Control that allows users on claude.ai and Cowork to select how much thinking effort Claude applies — from Low (faster, lower rate-limit usage) to Max. Opus 4.8 defaults to High effort for the best balance of quality and experience. According to Anthropic, Dynamic Workflows in Claude Code enables Claude to tackle massive tasks by planning, running hundreds of parallel subagents, and verifying outputs, supporting codebase-scale migrations across hundreds of thousands of lines of code. Additionally, the Messages API Update allows developers to insert system instructions mid-conversation without breaking prompt cache, and the minimum cacheable prompt length is reduced to 1,024 tokens from Claude Opus 4.7's higher threshold.
Standard pricing remains unchanged at $5 per million input tokens and $25 per million output tokens. As reported by Anthropic, Fast Mode for Opus 4.8, running at 2.5× speed, is priced at $10/$50 and is three times cheaper than previous fast modes. The model supports 1M token context window by default on the Claude API, Amazon Bedrock, and Vertex AI (200k on Microsoft Foundry) with 128k max output tokens. Fast mode is now available as a research preview on the Claude API with up to 2.5x higher output tokens per second. The model is available immediately across claude.ai, Claude API (claude-opus-4-8), and major cloud platforms.
