Skip to content
Submit incident
AII-1000-87489
Well documentedThree or more independent publishers report the core facts.Supporting sources are pending verification.

OpenAI Codex helps discover HTTP/2 Bomb DoS attack

OpenAI's Codex code generation model reportedly discovered an HTTP/2 Bomb denial-of-service attack that exploits compression mechanisms to crash web servers in seconds with minimal bandwidth. The technique targets known vulnerabilities in HTTP/2 implementations that have existed for decades, potentially affecting many web servers.

3 June 2026 Β· OpenAI Β· Security and cyberattacks
3 June 2026
AII-1000-87489
Quick facts
Organization
OpenAI (deployer)
Category
Security and cyberattacks
Date
3 June 2026
Drafted by
Raidu Incident Detection

What happened

OpenAI's Codex discovered a denial-of-service technique called HTTP/2 Bomb that exploits compression and flow-control stalling to knock web servers offline in seconds.

The attack can consume over 30GB of RAM within seconds using a single computer with a 100 Mbps connection, and researchers noted it targets decades-old vulnerabilities in HTTP/2 server implementations.

Reported impact

Major web servers were confirmed vulnerable to the HTTP/2 Bomb attack, which can render them offline before they can react.

Classification

Organization
OpenAI (deployer)
Category
Security and cyberattacks
Industry
Digital security
Harm status
Potential harm
Intent
Unknown
Timing
Unknown
Harm types
Economic or property
Harmed
Business
Failure modes
Hallucination
AI systems
OpenAI Codex

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident