In the 5Vs framework, what does variance measure?

In the 5Vs framework, what does variance measure?

• A. How standardized the work is; higher variance usually requires more flexible approaches and more human oversight.
• B. How easy it is to verify AI outputs.
• C. The potential harm if data is leaked.
• D. The speed of model training.

Answer: (A)

Variance in this framework captures variability or dispersion—how much things differ from one instance to another. When variance is high, work isn’t standardized; outcomes and tasks vary a lot, so you need flexible approaches and more human oversight to handle those differences and keep things reliable. If variance were low, processes would be more predictable and easier to automate. The other considerations—verification of outputs, security, or training speed—address different concerns and aren’t what variance specifically measures.

Variance in this framework captures variability or dispersion—how much things differ from one instance to another. When variance is high, work isn’t standardized; outcomes and tasks vary a lot, so you need flexible approaches and more human oversight to handle those differences and keep things reliable. If variance were low, processes would be more predictable and easier to automate. The other considerations—verification of outputs, security, or training speed—address different concerns and aren’t what variance specifically measures.