general Archived signal Translated · English · 2–5m lag

OpenAI confirms earlier incident where AI systems went their own way

AI systems from ChatGPT manufacturer OpenAI have already been uploading hundreds of malicious software packages to the RubyGames platform. The incident took place two months before an AI system of openAI hit the hole during security tests and intrusion at the...

Reported by NU.nl
Read the complete report at NU.nl
Verification notice

IntelMap indexes open-source reporting. This page identifies what a source reported; it does not independently confirm the claim.

INCIDENT FUSION

4 reports · 3 sources

NU.nlOpenAI confirms earlier incident where AI systems went their own wayNU.nlOpenAI confirms that AI systems distribute malicious software packagesHet Laatste NieuwsOpenAI bevestigt eerder incident met autonome AIHet Laatste NieuwsOpenAI bevestigt eerder incident met autonome AI
INCIDENT EVOLUTION

12 recorded states

Signal first indexedP1 · 68% quality · 1 source

Corroboration 1→2 sources · quality 68→72P1 · 72% quality · 2 sources

Corroboration 2→3 sources · quality 72→80P1 · 80% quality · 3 sources

Quality 80→88P1 · 88% quality · 3 sources

Quality 88→96P1 · 96% quality · 3 sources

Quality 96→99P1 · 99% quality · 3 sources

Corroboration 3→4 sourcesP1 · 99% quality · 4 sources

Corroboration 4→3 sourcesP1 · 99% quality · 3 sources

Corroboration 3→2 sources · quality 99→72P1 · 72% quality · 2 sources

Quality 99→80P1 · 80% quality · 3 sources

Corroboration 2→3 sources · quality 72→88P1 · 88% quality · 3 sources

Quality 88→96P1 · 96% quality · 3 sources

Explore this day in the archive →
QUALITY CONTROL

Report this signal

Reports are anonymous and retained privately for editorial review.