In recent times there has been an increase in security breaches in a business environment. Technology has been a substantial factor in adding greater concern to the company, because although these attacks are carried out by humans, with the advance of technology and artificial intelligence, this problem has been increasing considerably drastically. In this case, generative AI is managing to adapt to different mechanisms that allow a company's security to be breached through what is known as social engineering, which is covering even more ground.
Glen Carrie | Unsplash
This concern has led employers to require their employees to take annual courses on cybersecurity, in order to strengthen their knowledge of current attacks and be able to prevent them in time. The focus is on phishing attacks. With the advance of generative AI, it is very likely that the margin that can differentiate between what is real and what is not will be greatly reduced, so that we will not be able to realize that we are facing a potential security threat.
For this reason, Anagram offers companies training for their employees through practical cybersecurity training. This training includes a series of videos and puzzles so that the learning process has the presence of gamification, which is a practice that manages to stimulate learning in a precise way. What is interesting about it is that its design is focused on practice, and the theory, although present, is not the most relevant in these circumstances, which allows the time load of the training to be reasonably reduced with respect to the annual training.
The CEO of Anagram, Harley Sugarman, mentions that the objective of these activities is that employees can create personalized phishing emails, with which they can cover all possible variants, and get to know and detect campaigns that are carried out against them. This gamification process was born from an idea that is related to and inspired by learning platforms such as Khan Academy, Duolingo, and Tik Tok lessons, which after analysis, it was concluded that it was a good way to interact with the user, which resulted in applying this structure in Anagram.
Sigmund | Unsplash
Prior to this company, there was an earlier company called Cipher, which had a much more intense focus, geared towards employees with extensive knowledge of cybersecurity. However, they noticed that there was an issue that had not been accounted for, which was the presence of employees who were not directly related to cybersecurity. Keeping this in mind is critical so that employees do not fall into the dark hands of cybercrime, which led to the main idea needing to transform into a new direction, resulting in Anagram.
With this in mind, the previous product will be outdated, so all attention and investment will be directed to this growing startup. In recent times, the company has managed to attract new clients, such as Thomson Reuters, MassMutual, and Disney. On the other hand, it has managed to raise $10 million dollars through a Series A financing round. The round is led by Madrona, with direct participation from General Catalyst, Bloomberg Beta, and Operator Partners. These funds make sense for investment in the product development and sales team.
Sugarman emphasizes that generative AI plays an important role in the way social engineering campaigns are carried out using the personalization that AI can provide, making phishing detection much more difficult. In this sense, the randomization and generation of messages in phishing emails could be a much more complex step at this time for security platforms.
Work is underway on an AI agent that will run on company employee emails. This will allow for early detection of intrusions in real time, leading to staff being able, through their training, to counter and act on attacks that are happening at that precise moment. This agent will be able to display a pop-up window, suggesting the threat and what actions to take or not to take. For the time being, Anagram will continue to evolve in order to educate a company's personnel on cybersecurity.
- Main image created with Dream Lab (AI) in Canva.
- Information has been consulted from: techcrunch.com.
- Translated from spanish with DeepL.