This week, on March 27, the QA Financial Forum Chicago is taking place in the U.S. One of the event’s key speakers will be Clinton Sprauve, director of product marketing, application quality at Perforce, a provider of solutions to enterprise teams.
Sprauve plans to speak about leveraging AI for accuracy and clarity in testing, specifically testing complex financial charts and graphs.
Data-driven financial charts and graphics are vital to most financial firms conveying market insights too customers – but ensuring their integrity and real-time accuracy is challenging.
So how can we test the applications that are generating complex charts? “AI-driven validation with natural language prompts can enhance the testing of data-driven graphics, ensuring they reflect real-time, accurate financial information,” Sprauve explained.
“When you have a very complex chart or graph, it creates a bottleneck within that testing process. So what do most organisations currently do? Manual testing. Meaning they must sit there and view whether it’s supposed to look like that or they’re using screenshots, which is very unreliable,” he elaborated.
Instead, AI can easily take over, according to Sprauve. A bank that produces hundreds of charts and graphics on a daily basis, both for its customers as well as internal use, can now completely automatically check that the applications producing these charts are working correctly, using NLP.
How does this work? “NLP simplifies test creation by allowing users to write test scenarios in plain language,” Sprauve explained.
For instance, a user could input a prompt like “verify that the chart displaying US Equity Markets is correct” or is the Normalized Performance of SPX trending up or down in the graph”?
“The system can automatically generate the corresponding test. Also, it not only sees what is displayed on the screen but also understands the context and what is displayed,” he stressed.
A common fault it identifies could be if a chart displays incorrect data, such as showing the wrong values for the Nasdaq, S&P 500, Dow Jones indices, or a trending graph.
“In this case, AI Validation would recognise the request and cross-check the chart’s content with the expected values, flagging any discrepancies without human intervention,” he noted.
“AI is revolutionising how we test because it understands the visual and contextual aspect of the actual image that it is viewing.”
– Clinton Sprauve
The tool Sprauve’s company Perforce uses for complex charts and graphics is Perforce Perfecto, a cloud-based testing platform that powers AI-driven test automation for mobile, web, and API testing. It is already widely used by customers, including a rage of banks, to accelerate testing, reduce manual intervention, and ensure high-quality digital experiences.
“AI is revolutionising how we test, because it understands the visual and contextual aspect of the actual image that it is viewing,” he elaborated. “So it analyses not only what it sees but also understands what is underneath.”
According to Sprauve, the example leveraging AI for testing charts is an example of how QA teams should prioritise a new approach in future.
“Typically how teams test today is that they are trying to build better scripts using AI co-pilots,” he said.
However, “that’s just creating more work and will shrink productivity. The goal should be for AI to eliminate the need for scripting. … The goal should be autonomous testing,” Sprauve concluded.
Sprauve will be speaking at the QA Financial Forum Chicago on Thursday 27 March at 1.30. More information can be found here.
Why not become a QA Financial subscriber?
It’s entirely FREE
* Receive our weekly newsletter every Wednesday * Get priority invitations to our Forum events *
THIS WEEK
WATCH NOW
READ MORE
- Xceptor joins forces with OnCorps AI to target banks in automation drive
- QA Forum NY: Varo Bank’s Nicole Nunziata on secure coding
- Why QA teams may need to fear OpenAI’s bold AI agent plans
- Banks warned not to embed too much GenAI in digital systems
- UiPath turns to the cloud with its latest AI agent solution
