Revolut deployed the Codex tool and OpenAI models in software development and financial crime investigation
According to a case study published by the company OpenAI, the company Revolut cut code development time by 30-60% by deploying the Codex tool and OpenAI models and expanded automation of financial crime investigation.
The company Revolut integrated the Codex tool and models from OpenAI into its development process. According to a case study published by the company OpenAI, this allowed Revolut to cut code writing time by 30 to 60 percent.
According to the company OpenAI, the deployment was also meant to help expand the capacity for automated investigation of financial crime and increase the pace of innovation without, according to the stated claims, reducing the level of security. This is a vendor case study; the specific methodology for measuring the speedup and other technical details are not provided in the source.
See the source article for details.
Why it matters
The case study from the company OpenAI serves as a reference example for companies considering deploying AI assistants in software development and security/compliance processes in the financial sector. However, the stated figure of 30-60% coding speedup is a manufacturer's claim, not an independently verified result, so it should be treated with caution when planning your own investments.
Two audiences, two different impacts
What this means
For individuals
For individual developers, the case study shows that the Codex tool and OpenAI models can be genuinely incorporated into everyday coding with the stated speedup, which could be an incentive to try it yourself.
For a business
According to a case study, the company Revolut sped up its development process and expanded automation of financial crime investigation using the Codex tool and OpenAI models, which is a relevant reference point for companies considering deploying AI tools in development and compliance processes.
DevelopmentCheck the original
Event sources
only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.