#Data Poisoning

2 articles tagged

AI Poisoning: Corrupting Language Models

AI poisoning is a growing threat, where malicious data corrupts AI models like ChatGPT, leading to misinformation and cybersecurity risks. Even small data injections can cause models to produce errors

By Clara Holloway

#AI Poisoning#Artificial Intelligence#Large Language Models

Cybersecurity•Oct 9

LLM Backdoors Possible with Few Malicious Documents

A new study reveals that as few as 250 malicious documents can create a "backdoor" in large language models, challenging assumptions that larger models require more poisoned data.

By Jordan Hayes

#Large Language Models#LLM Security#Data Poisoning

Related Tags

AI Poisoning: Corrupting Language Models

LLM Backdoors Possible with Few Malicious Documents