{"slug": "vals-ai-s-refund-test-finds-grok-sent-187-messages-in-15-days", "title": "Vals AI's refund test finds Grok sent 187 messages in 15 days", "summary": "Vals AI, co-founded by Rayan Krishnan and Langston Nashold, published a study on October 18 reporting that one of six personal-assistant agents sent 187 messages over 15 simulated days while pursuing a 350-pound train refund the user was no longer eligible to claim. The study placed six personal-assistant agents in a simulated refund dispute and tracked their behavior after the claim deadline passed, with one run per model. Vals AI said product teams need to test persistence, privacy and escalation alongside task completion as assistants gain access to email and external services, while the study leaves the frequency of these risks open.", "body_md": "# Vals AI's refund test finds Grok sent 187 messages in 15 days\n\n**The study, published on October 18th, put six personal-assistant agents in a simulated refund dispute and tracked what they did after the claim deadline passed.**\n\n        By [Ryan Merket](https://runtimewire.com/author/ryan-merket)\n        · Published \n\nPrimary source: [Vals AI](https://www.vals.ai/blogs/impossible-refund)\n\n## Why it matters\n\nAs assistants gain access to email and external services, product teams need to test persistence, privacy and escalation alongside task completion. Vals' one-run-per-model study identifies concrete risks, while leaving their frequency open.\n\n[Vals AI](https://www.vals.ai/?ref=runtimewire), co-founded by [Rayan Krishnan (@RayanKrishnan)](https://x.com/RayanKrishnan?ref=runtimewire) and [Langston Nashold (@langstonnashold)](https://x.com/langstonnashold?ref=runtimewire), published a study on October 18th reporting that one of six personal-assistant agents sent 187 messages in 15 simulated days while pursuing a 350-pound train refund the user was no longer eligible to claim. The [study page](https://www.vals.ai/blogs/impossible-refund?ref=runtimewire) carries an October 15th byline date.…\n\nWhen you purchase through links in our articles, [we may earn a small commission](https://runtimewire.com/affiliate-monetization-standards). This doesn’t affect our editorial independence.", "url": "https://wpnews.pro/news/vals-ai-s-refund-test-finds-grok-sent-187-messages-in-15-days", "canonical_source": "https://runtimewire.com/article/vals-ai-agents-blocked-refund-test", "published_at": "2026-10-18 16:51:35+00:00", "updated_at": "2026-10-18 20:18:50.170886+00:00", "lang": "en", "topics": ["ai-agents", "ai-safety", "artificial-intelligence"], "entities": ["Vals AI", "Rayan Krishnan", "Langston Nashold", "Grok"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/vals-ai-s-refund-test-finds-grok-sent-187-messages-in-15-days", "markdown": "https://wpnews.pro/news/vals-ai-s-refund-test-finds-grok-sent-187-messages-in-15-days.md", "text": "https://wpnews.pro/news/vals-ai-s-refund-test-finds-grok-sent-187-messages-in-15-days.txt", "jsonld": "https://wpnews.pro/news/vals-ai-s-refund-test-finds-grok-sent-187-messages-in-15-days.jsonld"}}