![Voltaiq_Logo_Black_Transparent- 600.png]](https://1913583.hs-sites.com/hs-fs/hubfs/Voltaiq_Logo_Black_Transparent-%20600.png?height=42&name=Voltaiq_Logo_Black_Transparent-%20600.png){kind=logo}
Three Voltaiq AI use cases that give new perspective
Three prompts to get new perspectives from Voltaiq AI
Some teams get a lot more out of Voltaiq AI than others, and it usually comes down to the kind of questions they ask. Here are three new perspectives worth trying. Copy the prompts below and swap in your own data.
1. Ask it to poke holes in your conclusion
It's easy to trust the answer you hoped to find. That bias is hard to catch on your own, and it can sink a design review or a supplier meeting. So flip the script and ask Voltaiq AI to argue against you. Have it show the data that backs your idea and the data that works against it, so you spot your weak points before anyone else does.
Example prompt:
"I think these tests show that the capacity fade is primarily caused by electrolyte degradation. What is the evidence in the data that supports this conclusion? What are the potential arguments against this conclusion?"
You can also ask what data would prove you wrong, or have it make the best case for a different explanation.
2. Compare across different data sets
Most analysis stays within one program, batch, or cell. This can potentially hide patterns that show up only when you compare across groups. Voltaiq AI can compare experiments, production dates, cell types, and suppliers. Maybe a process changed between two builds, or a supplier only fails in the heat.
Example prompt:
"Compare yield and cycle life between cells from [supplier A] and [supplier B] tested at the same conditions."
3. Use statistics to check your results
With few samples or noisy data, the difference measured could be random chance. Voltaiq AI can help you determine the statistical significance. It can determine whether a difference is meaningful, whether you have enough samples to trust it, and which statistical test to use.
Example prompt:
"Is this new cathode material actually improving capacity fade, or could the difference just be random chance?"
You can also ask whether your sample size is big enough, or which statistical test fits your comparison and why.