GhostGPT: AI Tool for Cybercrime

GhostGPT is an AI system (a jailbroken LLM) whose misuse by threat actors is enabling business email compromise scams and malware creation. This represents direct, ongoing harm to victims via AI‐enabled phishing and cyberattacks, fitting the definition of an AI Incident.

The article describes an AI system (GhostGPT) being actively misused by criminals to create phishing campaigns and malware. This misuse has directly led to harmful outcomes by producing more convincing, higher-scale cyberattacks, qualifying as an AI Incident.

Hackers are directly deploying an AI system (GhostGPT) to generate complex malicious code that outsmarts detection and facilitates cyberattacks. This constitutes actual, realized harm driven by the AI’s outputs, so it is classified as an AI Incident.

GhostGPT is explicitly used by attackers to generate malicious content—phishing templates and malware code—leading to actual harms (successful phishing and ransomware campaigns). This constitutes an AI Incident, as the AI system’s use directly contributes to violations of rights and harms to communities.

The article describes a newly discovered AI system (GhostGPT) that is already being deployed by hackers to create malware and conduct phishing scams, directly facilitating harm. This constitutes an AI Incident because the misuse of the AI has realized harms to individuals and organizations through cybercrime.

The article explicitly identifies GhostGPT as an AI system (a chatbot likely based on a jailbroken large language model) used by criminals to create phishing emails and malware, directly facilitating cybercrime. This use leads to realized harms including scams and malware distribution, which constitute violations of rights and harm to communities. The AI system's development and use are central to these harms, meeting the criteria for an AI Incident rather than a hazard or complementary information.

GhostGPT is an AI system (a chatbot similar to ChatGPT) that is used maliciously to create malware and phishing emails, directly leading to harm such as fraud and cyberattacks. The article details how the AI's uncensored nature enables threat actors to bypass ethical safeguards and produce harmful content easily, which has already resulted in cybercrime incidents. The harms include violations of rights and harm to communities through scams and cyberattacks. This meets the criteria for an AI Incident because the AI system's use has directly led to realized harm.

GhostGPT is an AI system (a generative AI chatbot) whose use directly leads to harms including cybercrime, phishing, and malware creation, which cause harm to individuals and communities. The article details that the tool is actively sold and used by criminals, resulting in realized harm. Therefore, this qualifies as an AI Incident because the AI system's use has directly led to significant harms.