All corrections
1
Claim
It used to be that the worst thing these people could do was launch DDoS attacks against a particular target, or send spam email, or scrape for crypto and credit cards.
Correction

Historical botnets were already used for harms beyond DDoS, spam, cryptomining, and card theft, including credential theft and ransomware delivery.

Full reasoning

This sentence understates what botnet operators were already able to do before the AI scenario the post is discussing.

Official U.S. government sources document older botnets being used for banking-credential theft, broader credential harvesting, data exfiltration, and ransomware delivery — not just DDoS, spam, or scraping for cryptocurrency and credit cards.

Examples:

  • In 2014, the FBI said the GameOver Zeus botnet had as its "principal purpose" capturing banking credentials from infected computers, and that those credentials were used to initiate or redirect wire transfers.
  • In 2017, the Department of Justice said the Kelihos botnet was used for harvesting login credentials and installing ransomware and other malware.
  • In a 2021 joint advisory, CISA and the FBI said TrickBot was originally a banking Trojan and had evolved into malware capable of credential theft, data exfiltration, lateral movement, cryptomining, and dropping ransomware.

Because these documented botnets were already being used for ransomware and credential theft, the claim that the worst historical botnet operators could do was limited to DDoS, spam, or scraping for crypto/credit cards is factually incorrect.

3 sources
Model: OPENAI_GPT_5 Prompt: v1.16.0