Phishing scams have been a persistent issue for years, and for the most part, people have learned to avoid them by looking out for spelling and grammatical errors. However, the game has changed with the introduction of AI-based text generators like ChatGPT. This large language model chatbot developed by OpenAI can understand the meaning behind a question and has the potential to create sophisticated and convincing phishing emails that are difficult to spot.
How Fraudsters use ChatGPT
Writing phishing emails has always been a challenging task, especially for fraudsters who may not have strong language skills. But with ChatGPT, they can now create natural-sounding English responses customised to their targets. To generate a response, a fraudster simply needs to provide ChatGPT with a few prompts. For example, they could ask ChatGPT to inform Mrs Jones that she has won a prize and to provide her bank details to receive it. They could also ask for Mrs Jones’s passport information because the prize is a Mediterranean cruise. Finally, they could tell ChatGPT that Mrs Jones is suspicious and to adjust the text accordingly.
Implications of AI-Based Text Generators
Previous phishing attempts were much easier to detect because most spam filters were designed to identify commonly used phrases found in scam emails. However, with text generators like ChatGPT, fraudsters can create unique responses with natural-sounding language, making it almost impossible to identify the new generation of phishing emails. This poses a significant threat to individuals and businesses alike.
This new development highlights the need for more robust security measures and increased awareness of the potential threats posed by AI-based text generators like ChatGPT. The use of natural-sounding language and unique information makes it almost impossible to detect these types of phishing emails. As a result, individuals and businesses must be extra vigilant in protecting themselves against these sophisticated scams.
The Scale of the Problem in the UK
Phishing scams are a significant problem in the UK, with thousands of people falling victim to them each year. According to Action Fraud, the UK’s national fraud and cybercrime reporting centre, there were over 40,000 reports of phishing scams in 2020, representing a 20% increase compared to the previous year. The average loss to phishing scams in the UK is around £800, although some victims have lost much more. In total, losses to phishing scams in the UK amounted to over £200 million in 2020.
The financial losses associated with these scams are not the only concern. Phishing scams can also result in the theft of personal information, leading to identity theft and other types of fraudulent activities. This can have long-lasting implications for the victims, including significant financial losses and damage to their credit scores. It is essential for individuals to take steps to protect themselves from these scams and for businesses to implement measures to safeguard their employee’s and customers’ personal and financial information.
Protecting Yourself and Your Business from Phishing Scams
Individuals should be cautious when clicking on links in emails or text messages, especially if they are from unknown senders. They should also be wary of emails or messages that ask for personal or financial information, even if they appear to be from a legitimate source.
Businesses can take steps to protect themselves from phishing scams by implementing security training for employees, using multi-factor authentication, and regularly updating software and security systems. Reporting any phishing scams to relevant authorities is crucial in identifying the fraudsters and preventing them from targeting others.
The threat of AI-based text generators in phishing scams is significant, and individuals and businesses must remain vigilant. Checking for spelling and grammar errors is no longer enough to protect oneself from these sophisticated scams. If one falls victim, it is crucial to report it immediately to prevent further damage.
