For example, there are 3 areas that humans outdo robots. The first one is empathy. While building robots that can empathize with humans is the focus of much current research, keep in mind that neither scientists nor philosophers can agree on what emotions actually are.
Besides, no matter how sophisticated a robot’s emotional equipment becomes we still understand that it is merely a set of behaviors a machine has been programmed to sense and exhibit. As long as robots lack true empathy, complex human interactions will continue to require actual humans. The second one is flexibility. Robots are getting better at learning and applying their learnings to new situations, but they are still far from being able to respond to unexpected situations with the same resourcefulness as a person.
Most of the time, new situations still require human intervention and, when available and appropriate, new programming. And the third one acceptability and trust. We, humans, are gradually getting used to having robots in our environments. It will be awhile, however, before the average adult responds to a fully autonomous robot the way they do in futuristic movies — discussing an injury with a quasi-sentient machine as if it were fully able to comprehend the distress. And our most basic response to a non-human thing that can move around on its own is still, at the level of the reptilian brain, “Animal!”
For empathy, flexibility, acceptability, and more, there is still no replacement.