Emails have ruled as one of the top communication channels for a very long time, so it comes as no surprise that about 1.2% of all emails that are sent are malicious. This means about 3.4 billion malicious emails are sent out every day.
2023 is likely to witness even more phishing attacks, leading to millions of records getting stolen.
That said, there has never been a more important time to focus on empowering your security strategy and measures. Sounds simple enough, right?
But doing this just got harder, because of AI’s role in making these attacks more sophisticated.
What are AI-generated Phishing Emails?
Phishing usually involves a malicious party impersonating a business or individual in an attempt to retrieve personal and sensitive information such as credit card details, personal addresses, or passwords.
You may have encountered a phishing email attempt at least once and may even be adept at skirting around it, but phishing emails have now become even more sophisticated — thanks to Artificial Intelligence.
As an individual, you might come across an AI-generated phishing scam in the form of a private message or even a post on a public message board carrying a malicious link. As soon as you click on the link, the scammers will try to get access to your PII and sell this information to malicious parties.
You may even come across AI-generated phishing attempts in the form of a business email compromise (BEC). In this situation, the scammer might pose as your manager and spear phish employees to divulge banking details or get access to the office intranet or extranet.
Previously, you could tell phishing emails apart from legitimate ones by the way they were formatted or the error-laden content. But AI can write more polished, error-free and legitimate-looking copy giving you a false sense of security.
How to Identify Them?
So, the question now is — how can you identify such AI-generated phishing emails so that you can avoid falling prey to them?
Here are some of the distinguishing features that you should be aware of when identifying AI-generated phishing emails in your inbox.
- Use of Contextually Relevant Information: Some of the phishing emails might include information that is contextually relevant to assert its importance or credibility. For instance, it might reference a recent event related to your industry or a news item that might act as the perfect clickbait based on your interests and preferences.
- Improved Language: As discussed earlier, one of the tell-tale signs of a phishing email previously used to be the errors in the content. However, AI now helps scammers come up with grammatically correct and professional-sounding copy for emails making them seem more legitimate.
- Highly Sophisticated and Targeted Content: AI phishing emails can now leverage machine learning algorithms to analyse the likes, dislikes, habits and preferences of a specific individual. This makes it easier for scammers to come up with highly personalised messages, making it more difficult for you to distinguish them from legitimate emails.
For instance, you might receive an email from the eCommerce order fulfilment department of an online store that you frequently order from, with a tracking link to an order. In such cases, it is a good idea to cross-verify whether you have actually placed an order from the online store website or application, before clicking on any of the links in these emails.
- Deep faking Audio and Video: Cybercriminals and scammers often utilise deep fake audio and video as a part of their phishing attack along with the emails. For the uninitiated, deep fakes are audio or video clips manipulated with the help of AI impersonating an individual made to look or seem realistic. Deep fake audio is also used as a part of another variant of phishing, known as vishing wherein a malicious party may try to scam an individual or business through phone calls.
With AI providing access to advanced language models that barely ever make mistakes, phishing has become more difficult to detect. At this point, a genuine looking typo in the text might actually be more reassuring that the copy was written by a human as opposed to generated by AI.
Once you have identified specific patterns in the phishing emails that you are receiving, you should also consider creating a risk management register and keeping it updated.
3 Ways to Prevent AI-Generated Phishing Emails
Now that you know what AI-generated phishing emails are and how to identify them, let us look at some of the primary ways to prevent them from taking over your inbox. More importantly, these are steps that you can take to prevent yourself from falling victim to one of the phishing scams.
1. Regular and Frequent Clean Up
One of the major problems causing a phishing issue are those idle applications and plugins that connect to the internet and keep downloading data. With that in mind, it helps to let your employees know that they should keep their digital assets – online and offline, clean from time to time. There are many ways to conduct risk assessments using due diligence tools, which can help you identify risks in a timely manner.
Whether you are getting phishing emails on your personal or business email account, you will need to clean up your inbox and get rid of suspicious emails periodically. These clean ups will need to be frequent so that you can minimise and eliminate potential damage that can be caused by phishing emails.
2. Simplify The Process of Reporting Phishing
To prevent any major damage from an AI-generated phishing attack, you will need to have an advanced alert system. As a business, you should enable your IT team to track and analyse AI campaigns so that you can inform your employees about anti-phishing technologies and alert them in a timely manner.
If you already have business process automation measures in place, you have even more reason to automate reporting and simplify reporting of phishing emails as much as possible. Plus, you should make it as simple as possible for employees to report phishing emails, and have mechanisms to capture specific meta data and identifying information from the emails to be forwarded and reported.
As a result, governments have enforced the usage of DMARC (domain-based message authentication, reporting, and compliance) which enables you to utilise several mechanisms and policies like the ones we have discussed so far. Having a strong and comprehensive phishing report is one of the indicators of a strong security infrastructure.
3. Have a Multi-Layered Security Approach
Even with the most comprehensive and efficient malware at your disposal, you will need to get insights into an individual’s headspace to be able to come up with the right security measures to protect your business from phishing attacks. A multi-layered security strategy is the perfect way to go beyond conventional cybersecurity measures to prevent phishing.
Some of the main components in a multi-layered security strategy include a human firewall, AI-based security technology, highly upgraded and top-quality authentication measures, AI policies and procedures, and consistent employee training for security awareness. A marriage of all these safeguards in your strategy will ensure that you are in a better position to prevent any damage from a phishing attack.
While phishing has been a popular tactic used by cybercriminals around the world for a long time, these attacks are now evolving, thanks to artificial intelligence. So, you need more comprehensive and sophisticated security measures in place to tip the odds in your favour. But to be able to prevent a phishing attack in a timely manner, you will need to identify them on time and take appropriate security measures.
Carl Torrence is a Content Marketer at Marketing Digest. His core expertise lies in developing data-driven content for brands, SaaS businesses, and agencies. In his free time, he enjoys binge-watching time-travel movies and listening to Linkin Park and Coldplay albums.
Twitter – https://twitter.com/torrence_carl
LinkedIn – https://www.linkedin.com/in/torrence-carl/