By now you've probably heard of Citi's "near-miss" of an $81tn (yes, "trillion") internal transfer:
This is when you’ll expect me to go on about processes and risk controls. And I will do just that.
I'll start with this excerpt:
Citi's technology team instructed the payments processing employee to manually input the transactions into a rarely used back-up screen. One quirk of the program was that the amount field came pre-populated with 15 zeros, which the person inputting a transaction needed to delete, something that did not happen.
(Source: FT, "Citigroup erroneously credited client account with $81tn in ‘near miss’")
To loosely paraphrase safety expert Todd Conklin: when designing a system, make it easy to do things the right way, and hard to do things the wrong way.
If you are creating tools for people, a suitable UI/UX on the front end is just as valuable a risk control as testing and checks on the backend. You need both.
(So … Whatever money you are spending on UI/UX, spend more. And listen to your designers.)
Unreliable chatbots
The BBC ran the numbers on chatbot goofs
Getting the bots to do the work
This is really a lesson in tech interview practices