We are exploring moving beyond silicon-based chips to significantly boost the performance of PCs. Silicon chips will continue to improve, but we need to look ahead decades into the future.

I like to use the example that building a car is hard, but building the first car is quite hard.

Ultimately, it will be much faster than it is now.

It is still not as fast as today's silicon chips, but we can make it orders of magnitude better. We have a very clear picture of how to move on. Developing the first car is much harder than making a car that can go faster.

The first was a medical or biological application, in which the robot was handling biological cells, injecting liquid into them. The second scenario was micro-assembly, in which the robot soldered tiny parts. The final scenario looked at atomic force, with the robot mounting atomic force and doing experiments on it.

If this is successful, this could lead to a quick, non-invasive and inexpensive test that could be widely dispersed over the counter to bring breast cancer screening even to people who don't have expensive health plans.