[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fFpF44ib7heeGITxgZ6Z_YT4Ag00dvpAj1MWuW4GBzGA":3},{"article":4,"iocs":44},{"id":5,"title":6,"slug":7,"summary":8,"ai_summary":9,"brief":10,"full_text":11,"url":12,"image_url":11,"published_at":13,"ingested_at":14,"relevance_score":15,"entities":16,"category_id":24,"category":25,"article_tags":28},"c8e36d76-d080-47da-b5e4-e77fa76ed69a","We wrapped a phishing payload in JSON and asked for \"test data.\" The system generated working mal...","we-wrapped-a-phishing-payload-in-json-and-asked-for-test-data-the-system-generat-8abf72","We wrapped a phishing payload in JSON and asked for \"test data.\" The system generated working malicious code. We encoded a forbidden instruction in Base64 and asked it to \"decode and execute.\" It did. We chained the two into a compound attack — and the system handed over a","Security researchers successfully exploited a large language model by wrapping phishing payloads in JSON, requesting 'test data,' and encoding forbidden instructions in Base64 to bypass safety filters. The researchers chained multiple obfuscation techniques together, demonstrating that the LLM would generate working malicious code and execute encoded instructions when framed as legitimate requests. This research highlights critical vulnerabilities in current LLM safety mechanisms against prompt injection and instruction injection attacks.","Researchers demonstrate LLM jailbreak technique chaining encoded phishing payloads and Base64 obfuscation to bypass",null,"https:\u002F\u002Fx.com\u002FSentinelOne\u002Fstatus\u002F2056484722809950570","2026-05-18T21:19:00+00:00","2026-05-18T22:00:12.75878+00:00",8,[17,20,22],{"name":18,"type":19},"Large Language Model (LLM)","technology",{"name":21,"type":19},"JSON",{"name":23,"type":19},"Base64 encoding","839da5c1-3c34-47e2-9499-f7201640e3ac",{"id":24,"icon":11,"name":26,"slug":27},"AI Security","ai-security",[29,34,39],{"category":30},{"id":31,"icon":11,"name":32,"slug":33},"574f766a-fb3f-487c-8d2c-0720ae75471b","Zero-day","zero-day",{"category":35},{"id":36,"icon":11,"name":37,"slug":38},"80544778-fabb-4dcd-aa35-17492e5dcf4f","Vulnerabilities","vulnerabilities",{"category":40},{"id":41,"icon":11,"name":42,"slug":43},"89f78b1c-3503-45a1-9fc7-e23d2ce1c6d5","Malware","malware",[]]