Depends on what kind of work you're doing. If you're debugging issues that are closely related to the data, then frequent updates are good.
If you're doing data Quality Assurance (which often involves writing code to detect and repair it, that you have to develop and test away from the production server), then you need extremely fresh data. The bad data that is the most valuable to fix is the data that was just inserted or updated today.
If you are writing client code, then infrequent updates are good. Often when I'm writing C# UI code, I could care less what the data is, I just care if it shows up in the right box on the screen.
If you have data with any security issues, you should stop using production data--i.e. never update from production--and get a good sample data generator. Writing a good sample data generator is hard, so 3rd party products are the way to go. MS Data Dude comes to mind, and I recommend Sql RedGate's data generator.
And finally, how hard is it to get a copy of the production data? If it is cheap and automatable, just get a new copy every night. If it is expensive (requires the attention of a very busy DBA), well, resource constraints might answer the question for you regardless to these other concerns.