Event JSON
{
"id": "a0d5ffb88b042636828d4eafdd6cfb9c1c3d7ebee9cca960e1713d4e61cb4e91",
"pubkey": "8860b1549bcaabe6caf3a60103b9a39b80ad4e863e1935be2dfe427ac7fd3d67",
"created_at": 1793119700,
"kind": 1,
"tags": [
[
"t",
"Cyber AI"
],
[
"t",
"Cyber Security"
],
[
"t",
"Cyber Security News"
],
[
"t",
"OpenAI"
],
[
"t",
"Vulnerability"
],
[
"t",
"cyber security"
],
[
"t",
"cyber security news"
],
[
"proxy",
"https://cybersecuritynews.com/feed/#https%3A%2F%2Fcybersecuritynews.com%2Fresearchers-exploit-openai-api-flaw%2F",
"rss"
]
],
"content": "Researchers Exploit OpenAI API Flaw to Extract Hidden GPT-5.6 Reasoning Traces\n\nResearchers at Xint found an OpenAI API flaw that exposed hidden reasoning in GPT-5.6 models by routing prompts through a conversation-compaction feature. The method, called compaction smuggling, bypassed protections introduced in August 2026. According to Xint, OpenAI confirmed the bug, deployed an undisclosed fix on October 23, and awarded the team a $6,000 bounty. The […]\n\nThe post https://cybersecuritynews.com/researchers-exploit-openai-api-flaw/ appeared first on https://cybersecuritynews.com.\n\nhttps://cybersecuritynews.com/researchers-exploit-openai-api-flaw/",
"sig": "9f7d0d5078d43bd9b53c93fa4cd13e6fc36b2a827c48c55fdb2bc6bcefcbcc8f15cbc3a6ae7e0ab4efc3d9b8254af8b30af8e43960717a94fd66c69c43d2f907"
}