Another success of AI is being able to write better phishing emails than humans.
Natural Language Processing or Artificial Intelligence is working unexpectedly in various fields. This time, AI has done an unexpected job in phishing emails. Researchers have discovered that deep learning languages can be used to create emails for phishing campaigns.
Researchers have long debated whether scammers could use machine learning algorithms to write phishing messages.
Last week, a presentation was made by the Singapore government’s technical agency at the Black Hat and Defcon Security Conference in Las Vegas, USA, where about 200 people were sent their own phishing emails and AI-generated phishing mails.
Some links were attached to the message, which was not malicious. After clicking on it, the information was prepared to reach the researcher. Surprisingly, most of the people who received the message clicked on the link sent by AI rather than the link sent by them.
According to the agency, training AI for this requires some expertise in the beginning, which will cost billions of rupees, but once it is developed as an AI as a service, it will do a lot of work at a low cost. No coding is required for this. Once the data is input into the machine, it will start working automatically.
The agency said it was only the first stage of the investigation, as it was conducted in small numbers and among people it knew. The agency says more research is being done on how AI can help with phishing and spare phishing. However, many questions are being raised about the development of Open AI in human existence.
It is said that research is being done to reduce the impact of using AI in such cyber crimes on society.