How Stripe Uses GPT-4

Artificial Intelligence (AI) has been making waves in various industries, and the payment platform sector is no exception. Stripe, a leading payment platform, has embarked on an innovative journey, leveraging the capabilities of OpenAI‘s latest language learning model, GPT-4, to enhance its features and functionality.
Earlier this year, Stripe embarked on an unconventional project. They asked 100 employees to pause their regular duties and instead, brainstorm ways to improve the payment platform using GPT-4. This diverse group of engineers, hailing from teams such as support, onboarding, risk, and documentation, were tasked with identifying areas where AI could enhance Stripe’s features or workflows.
Eugene Mann, the product lead for Stripe’s Applied Machine Learning team, explained that their mission was to identify products and workflows across Stripe that could be accelerated by large language models. The access to GPT-4 opened up a world of possibilities, revealing a host of problems that could be solved effectively with the AI model.
Stripe uses GPT-4 to provide users with a better experience and prevent fraud
Stripe’s primary users are developers who integrate Stripe’s software into their own projects. The more proficient these developers become at deploying Stripe, the more Stripe can expand its reach in the digital payments universe, thereby growing the GDP of the internet.
Stripe had previously been using GPT-3 to assist their support team in tasks such as routing issue tickets and summarizing user queries. However, with GPT-4, they were able to explore even more applications. After testing 50 potential applications, 15 prototypes were identified as strong candidates for integration into the platform. These included support customization, answering support questions, and fraud detection.
One of the key ways Stripe aims to improve user experience is by understanding exactly how each business uses its platform. This allows them to customize support accordingly. However, this process can be time-consuming. To address this, Stripe now uses GPT-4 to scan websites and deliver summaries, which have proven to be more accurate than those written by humans.
Another area where GPT-4 has proved invaluable is in answering support questions about documentation. Stripe provides extensive technical documentation to support developers. GPT-4 can digest and understand this documentation, acting as a virtual assistant to identify relevant sections and summarize solutions.
In addition to enhancing user support, GPT-4 has also been instrumental in managing malicious activity. By analyzing the syntax of posts in forums like Discord, GPT-4 can flag accounts for Stripe’s fraud team to investigate, helping to maintain the integrity of the community.
Looking ahead, the Stripe team is considering further applications for GPT-4. One idea is to deploy it as a business coach that can understand revenue models or advise businesses on strategies. As GPT-4 continues to evolve, the potential applications are only broadening.
In conclusion, Stripe’s innovative use of GPT-4 is a testament to the transformative power of AI. As Mann puts it, “Some of these features feel like magic.” This journey is a clear demonstration of how AI can be harnessed to solve real-world problems and enhance user experience, paving the way for a future where AI is an integral part of our daily lives.
LAStartups.com is a digital lifestyle publication that covers the culture of startups and technology companies in Los Angeles. It is the go-to site for people who want to keep up with what matters in Los Angeles’ tech and startups from those who know the city best.