OpenAI has released a new threat intelligence report revealing how cybercriminals and coordinated fraud networks are exploiting ChatGPT to power online scams, impersonation campaigns, and deceptive business operations.

According to the report, malicious actors have used AI-generated content to craft convincing romance scam messages, pose as legal professionals, draft fraudulent documents, and automate large-scale outreach campaigns. The company says these schemes are becoming more polished and harder to detect due to the fluency and adaptability of AI-generated text.

Rise of AI-Powered Romance and Impersonation Fraud

One of the most prominent abuses highlighted involves dating and relationship scams. Fraudsters reportedly used ChatGPT to generate emotionally manipulative messages designed to build trust with victims over time. AI tools allowed scammers to maintain consistent, believable conversations across multiple targets simultaneously.

In other cases, individuals posed as fake lawyers or legal consultants, using AI to draft official-looking contracts, legal notices, and advisory emails. The report suggests that generative AI significantly lowers the barrier for criminals to produce professional-grade content without specialized expertise.

Coordinated Influence and Outreach Campaigns

Beyond financial scams, OpenAI identified networks leveraging ChatGPT for influence operations and coordinated messaging campaigns. These groups allegedly used AI to generate social media posts, comments, and persuasive narratives aimed at amplifying specific viewpoints or misleading audiences.

While the company did not disclose all operational details, it confirmed that multiple accounts connected to such campaigns were banned after internal investigations.

Detection, Disruption, and Account Bans

OpenAI stated that its safety teams actively monitor suspicious usage patterns and collaborate with external partners when necessary. The report outlines how investigators track behavioral signals, infrastructure links, and content patterns to identify coordinated abuse.

Once confirmed, accounts are suspended and, in certain cases, referred to relevant authorities. OpenAI emphasized that the majority of ChatGPT usage remains legitimate, but acknowledged that preventing misuse at scale remains an ongoing challenge.

Strengthening Safeguards and Monitoring Systems

In response to evolving threats, the company says it continues to refine automated detection systems and introduce stricter verification layers. OpenAI also highlighted investments in threat intelligence research, red-teaming exercises, and improved abuse-reporting mechanisms.

The report underscores a broader industry issue: as generative AI tools become more accessible and powerful, so do the opportunities for malicious exploitation.

Balancing Innovation and Responsibility

OpenAI concluded that transparency around misuse is essential to maintaining public trust. By publicly documenting emerging abuse patterns, the company aims to help policymakers, cybersecurity professionals, and the broader tech ecosystem respond more effectively.