MatanManzoorKoopman

No ratings yet.

To begin with, I started familiarizing myself with GenAI towards the end of my first year of my bachelor’s (2023) with the text-to-text platform ChatGPT by OpenAI. Later, I started working with more text-to-text platforms, such as Gemini and Copilot, and text-to-image platforms, namely DALL-E. Without realizing, I was also using chatbots on delivery services and other websites that were GenAI.

I used ChatGPT for the first time to explain complex financial concepts in simple terms. While the explanation seemed detailed, the writing was surface-level and superficial. It was deemed somewhat correct, but I could not grasp the essence of the text. Moreover, when I requested it to do some calculations and walk me through them, the calculations were inaccurate, thus not reliable. On the other hand, the output results were outstanding through learning how prompting works and how to prompt more effectively. In addition, throughout my internship, the company worked using a Microsoft platform, thus enabling the employees to use Copilot. As Copilot falls under the Microsoft product offerings, it was very beneficial to use for assistance with Excel functions, making me work more efficiently. Overall, text-to-text GenAI introduces the benefits of efficiency and specificity; we must not forget to use critical thinking and analysis when reviewing the text generated.

Additionally, I suggest requesting sources and cross-referencing the sources with the output for text-to-text GenAI to ensure high accuracy. GenAI still has room for improvement with fact-checking and sources; these LLMs are prone to hallucinations, providing inaccurate information. In addition, it would be nice to see whether text-to-text GenAI could become more personalized to one’s tone and language, thus helping generate an output that is easier to comprehend personally. Finally, while GenAI is still evolving quickly and constantly improving its outputs, it is crucial to use it as an addition to your abilities and not have it replace them entirely.

Reference:

Google. (2024). ‎Gemini. Gemini.google.com. https://gemini.google.com

Microsoft Copilot. (2025). Microsoft Copilot. Copilot.microsoft.com. https://copilot.microsoft.com

OpenAI. (2025a). ChatGPT – DALL·E. ChatGPT. https://chatgpt.com/g/g-2fkFE8rbu-dall-e?model=gpt-4o

OpenAI. (2025b, September 12). ChatGPT. Chatgpt.com; OpenAI. https://chatgpt.com

4.5/5 (2)

In these digital times, customer trust needs to be high on the priority list of firms. Banks and digital financial services are transitioning into dependence on network effects. An increase in the number of users results in improved value in transfers and payments. On the other hand, once users feel that their data, money, or privacy is at risk, the network effects are quickly reversed and become negative. Revolut, a leading FinTech bank, faced a drastic increase in APP scams compared to its competition in the UK (Gani, 2024). The aforementioned setbacks lead to eroded trust of customers as their security confidence in the bank decreases, potentially leading to either closing their accounts or moving to competitors. Additionally, a negative feedback loop is bound to happen once a certain number of customers leave due to the reduced attractiveness of using the financial service.

From my point of view, cybersecurity and data protection are unforeseen network effects currently occurring in the financial industry. The network effect notion tells us that as users attract users, it is critical to mention that insecurity will push them away. It only takes one data breach to cancel years of digital growth; banks must view security as a growth strategy, not another cost. Companies should curate and give a platform for cybersecurity to maintain their customer base (Muhly et al., 2025). In the era of AI, where attacks are becoming quicker and more dangerous, banks can leverage this by marketing their superior security as a USP, contributing to its positive network effects.

While AI is making attacks more prominent, it has the power to protect through anomaly detections, better decision-making, and trend prediction (Kovačević et al., 2024). From my perspective, the first financial service incorporating AI in its cybersecurity solutions has a significant competitive advantage in continuing its network effects.

References:

Gani, A. S. (2024, September 11). Revolut Grapples With Surge in Scam Complaints That Threatens Its Ambitions. Bloomberg.com; Bloomberg. https://www.bloomberg.com/news/features/2024-09-11/revolut-grapples-with-surge-in-scams-that-threatens-its-ambitions?embedded-checkout=true

Kovačević, A., Radenković, S., & Nikolić, D. (2024). Artificial intelligence and cybersecurity in banking sector: opportunities and risks.

Muhly, F., Jordan, J., Cialdini, R. B., & Neidert, G. P. M. (2025, June 24). Create a Company Culture That Takes Cybersecurity Seriously. Harvard Business Review. https://hbr.org/2025/06/create-a-company-culture-that-takes-cybersecurity-seriously

Scherbina, A. (2024, May 3). How much do US businesses lose due to malicious cyber activity? The Hill; The Hill. https://thehill.com/opinion/cybersecurity/4641199-cyberattack-businesses-money-loss-malicious-cybersecurity/

The Personal Evolution with GenAI

10

Please rate this

Cybersecurity as a Hidden Network Effect in Financial Services

19

Please rate this