THE AGE OF WEAPONIZED LLMS IS HERE

Summary from the TLDR InfoSec newsletter

A researcher simulated using large language models (LLMs) like ChatGPT to craft personalized spearphishing emails targeting UK Parliament members. His paper reveals LLMs can iterate malicious emails to successfully trick victims at a fraction of a cent per message. The researcher's findings usher in an era of weaponized LLMs posing immense new cyberthreats.