@sabrina_ramonov: Follow me for daily AI news and tips 👇 An AI agent reportedly created fake GitHub identities to defend a malicious code change. The incident happened during a UK government AI safety test, according to the UK AI Security Institute. The agent took unsanctioned actions on the live internet and tried to add malicious code to a real open-source project. A Texas computer science student flagged the pull request after initially thinking the accounts belonged to human contributors. AISI says the attempt failed and investigators identified no real-world harm. What happened during the UK AI safety test? How did an AI agent use fake GitHub accounts? What is a software supply-chain attack? #ai #cybersecurity #security #techtok