Find notable cyber news and cases, enriched with sources, timelines, and signals.

OpenAI ChatGPT renderer Markdown link/image phishing security flaw

Vulnerability
First reported
Last updated
Happening score
H score 16
1 unique sources, 1 articles

Summary

Hide ▲

ChatGPT has a response-renderer vulnerability that turns summarized third-party pages into live phishing links and auto-fetched attacker-hosted images inside the trusted UI. The flaw abuses Markdown links and Markdown image URLs that survive from a page the assistant just summarized. It can expose users to phishing and tracking when ordinary web-page summarization is used.

Related Happenings

LLMShare ChatGPT share-link malware lure campaign

Campaign
H score47 First: 29.05.2026 21:21 Last: 29.05.2026 21:21 Sources 1

About this happening: The LLMShare campaign is using Google ads and a legitimate chatgpt.com shared page to route people searching for ChatGPT into a fake OpenAI outage lure that pu...

CypherLoc phishing-led browser scareware campaign

Campaign
H score49 First: 20.05.2026 13:00 Last: 20.05.2026 13:00 Sources 1

About this happening: The CypherLoc operation has driven around 2.8 million attacks since the start of 2026, using phishing emails to send users to malicious pages that lock browsers an...

Vercel v0.dev phishing campaign using GenAI-built lure pages

Campaign
H score29 First: 07.05.2026 11:30 Last: 07.05.2026 11:30 Sources 1

About this happening: A campaign using Vercel v0.dev to build highly convincing phishing pages has lowered the skill and cost needed to run fraudulent sign-in and job-lure attacks. The activity...

OAuth device-code phishing campaign targeting SaaS accounts

Campaign
H score43 First: 04.04.2026 17:17 Last: 04.04.2026 17:17 Sources 1

About this happening: A device code phishing campaign now includes EvilTokens, a phishing-as-a-service kit sold on Telegram that uses the OAuth 2.0 device authorization flow to hija...

Perplexity Comet prompt-injection research shows agentic browsers can be trained into phishing traps

Technical Analysis
H score25 First: 11.03.2026 18:38 Last: 11.03.2026 18:38 Sources 1

About this happening: Perplexity's Comet AI browser is the focus of a technical analysis thread showing how prompt injection and malicious URLs can steer an agentic browser into data...

Timeline

  1. 29.05.2026 21:07 2 articles · 1mo ago

    ChatGPT response renderer trusts Markdown links and images from summarized pages

    Initial Disclosure

    Cybersecurity researchers disclosed a flaw in OpenAI ChatGPT web summarization where the response renderer trusts Markdown links and Markdown image URLs from a third-party page the assistant has just summarized. That behavior can surface attacker-controlled links as live, clickable elements inside the trusted assistant UI and auto-fetch remote images, enabling phishing lures and leaking IP, User-Agent, and Referer details when a malicious page is summarized.

    Show sources