Can AI Write a More Convincing Phishing Email Than Humans?
Can AI produce more efficient phishing emails than humans?
NLP and effective translation techniques become an invaluable tool for fraudsters. What should one know about the role of AI in cyberattacks?
How AI Writes Phishing Emails
AI can be used by hackers to make more efficient phishing material. With ChatGPT it became easier to produce content. A fraudster could load into AI a ton of emails and request for an original one. AI can understand and produce prudent content.
In a perfect world, AI creates an email, that is 100% similar to a human-witten one. The fraudster can specify then different details about a company or a person. AI then translates the message into different languages and makes the message more personalized, which prevents it from being sent to spam folder.
Evidence suggests that AI-based messages are of the same efficiency as human. However, the good news is that hackers typically have limited access to programs like OpenAI.
AI is a great tool for hackers due to its efficiency (fast scam creation and variability) and language. AI can also target people anywhere in the world, since it’s pretty accessible. So, the good news is that it can be as efficient as human-produced emails, but unlikely more efficient than those.
Here are some tips on Internet security: using cloud storage ( it’s not connected to any physical device and it's a challenge to get rid of it or harm it, for instance). Also, customers can run vulnerability checks. People can also make a code system for trusted correspondents like family, friends, colleagues. A person from the trusted circle may write a code or a phrase for identification, where the message comes from. The message should be something very uncommon for AI to be aware of.
Computer specialists may utilize machine learning for observation of natural speech patterns of lawful emails. AI may be taught unique writing styles of individuals and then recognize fake emails.
However, it is possible for machines to produce believable bogus emails. The problem for hackers is that they don’t typically have technical resources to teach AI to imitate an individual’s handwriting efficiency.
Despite the fact, AI is a proven powerful tool for phishing email creation, there is not enough evidence to suggest that AI-written emails are more effective. The red flag comes with messages with call to action - but those are still in trend, no matter who wrote them, a human or a machine.
AI Catalog's chief editor
Share on social networks:
