Fraudsters are always finding new ways to get their notifications into your inbox. While spam filters have become more advanced, they often don’t keep up with the constantly changing tactics of those who wish to defraud unsuspecting people.
ChatGPT has made a revolutionary breakthrough in the world of AI and unfortunately, it could be misused by spammers to trick people into clicking malicious links, buying products, or giving away personal information. Such technology could also help them bypass anti-spam filters.
Why should I be concerned?
Although spam emails have long been a nuisance to internet users, with the emergence of AI-generated emails, this threat has become even more sophisticated and persuasive as AI is capable of understanding the personal preferences, principles, and idiosyncrasies of people.
AI-generated spam emails are designed to mimic human communication and use a variety of psychological triggers to manipulate the recipient into taking a specific action.
Although they may appear harmless at first glance, these emails can be very convincing and lead to serious consequences if one is not careful.
So, what is spam?
Spam emails, messages, and phony reviews on goods all come under the umbrella of unsolicited communication from an unidentified source. This is a form of digital intrusion that should be avoided at all costs as it can tarnish your online reputation. Spammers try to persuade people to take specific actions, such as buying a product, clicking on malicious links, downloading malware, or giving away personal information such as social security numbers, passwords, and credit card details.
Is spam profitable?
Believe it or not, spam is profitable. Just one email blast can make US$1,000 in just a few short hours. The cost of this to spammers is only a few dollars – excluding the setup cost. If someone were to run a pharmaceutical spam campaign online they could expect to earn around $7,000 per day on average.
Spammers don’t have access to mailing lists that have been opted-in by users. They use various deceptive techniques, such as the well-known “Nigerian prince” scam, where a Nigerian prince asks for help in unlocking a significant sum of money and promises to award you generously in return.
Digital natives generally no longer fall for such scams, however, the outrageousness of these requests still may be attractive to those who lack tech knowledge or are advanced in age. As a result, this method allows them to filter out the people who have a high likelihood of being scammed.
So, internet-savvy folks who are wary of scammers have nothing to worry about, right?
Wrong – AI opens up new opportunities for spammers as they can obtain more personal data and tailor their messages to be even more persuasive. Unfortunately, this kind of AI-powered spam could become more difficult to detect and counter, thanks to its ability to use social media posts and other sources of information.
Spammers can now exploit the limited data they have to predict more about people. For example, with only a LinkedIn page and posts, as well as a profile picture or two, malicious spammers with LLM (Linked-In Lookalike Model) could estimate one’s political alignment, marital status, or even life objectives.
And spammers are increasingly persistent in finding ways to get past our filters, making us more likely to click a link or even interact with them. This gives them a higher chance of using their customized tactics to persuade us further. LLMs can make a huge difference here. Initial results show that LLMs can be utilized effectively to put forward persuasive arguments in areas ranging from politics to health policy.
When something looks suspicious, I hover over hyperlinked text or an image in the email to view the landing page domain. If things don’t add up, I don’t click. Check out the spam link below. While Instagram is a legitimate site, its web address obviously doesn’t start with “I.”
Oftentimes, you’ll find the landing page address to be something totally inconceivable like www.wyz.987654.com.
The perspective with which we view the introduction of new technologies is determined by those who create and wield them. They can be seen as a source of either great benefit or considerable harm, depending on how they are managed.