{"data":{"id":"7991749a-5c67-41a6-b5e0-132daca7d9bf","title":"Poison Claude Sells Discounted Claude Access While Its Operator Sees Every Customer Prompt","summary":"Cybersecurity researchers discovered illegal services like Poison Claude that sell discounted access to Anthropic's AI models by exploiting free AWS credits and routing user requests through their servers. A major privacy risk is that these proxy services can see all customer prompts and inputs, since they must forward them to the actual AI model to get responses back.","solution":"A configuration error exposing Poison Claude's API status endpoint 'api.claudeopus[.]shop/api/status' has since been fixed. Following responsible disclosure, Cloudflare placed a phishing warning in front of the main Poison Claude domain, though it declined to take action on the API domain itself.","labels":["security","privacy"],"sourceUrl":"https://thehackernews.com/2026/08/poison-claude-sells-discounted-claude.html","publishedAt":"2026-08-05T15:36:03.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"high","attackType":["supply_chain","data_extraction"],"issueType":"news","affectedPackages":null,"affectedVendors":["Anthropic","OpenAI"],"affectedVendorsRaw":["Anthropic","Claude","Opus 4.8","Opus 4.7","Opus 4.6","Sonnet 4.6","OpenAI","GPT Codex 5.5","DeepSeek","Moonshot AI","MiniMax"],"classifierModel":"claude-haiku-4-5-20251001","classifierPromptVersion":"v3","cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"patchAvailable":null,"disclosureDate":"2026-08-05T15:36:03.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"moderate","impactType":["confidentiality","integrity"],"aiComponentTargeted":"api","llmSpecific":true,"classifierConfidence":0.95,"researchCategory":null,"atlasIds":null}}