New technology now allows for artifical skin that is sensitive to mechanical deformation, temperature and wetness  [1]. From the article:

Sensory receptors in human skin transmit a wealth of tactile and thermal signals from external environments to the brain. Despite advances in our understanding of mechano- and thermosensation, replication of these unique sensory characteristics in artificial skin and prosthetics remains challenging. Recent efforts to develop smart prosthetics, which exploit rigid and/or semi-flexible pressure, strain and temperature sensors, provide promising routes for sensor-laden bionic systems, but with limited stretchability, detection range and spatio-temporal resolution. Here we demonstrate smart prosthetic skin instrumented with ultrathin, single crystalline ​silicon nanoribbon strain, pressure and temperature sensor arrays as well as associated humidity sensors, electroresistive heaters and stretchable multi-electrode arrays for nerve stimulation. This collection of stretchable sensors and actuators facilitate highly localized mechanical and thermal skin-like perception in response to external stimuli, thus providing unique opportunities for emerging classes of prostheses and peripheral nervous system interface technologies.

Whom it may concern?

When the ultimate goal of this type of prosthetic skin is to enable amputees to feel various types of external stimuli, as the paper states, then of course you have me interested in whether you successfully convinced someone to lend their arm, or what else it was that was done to get this on the road. While the research clearly targets arm amputees as potential recipients of these new things, the Korean based paper only mentions animal experiments in their method section, while explaining some results of tracking hands of wrists obviously of human subjects that are unknown from the method section. We thus are left in the dark about what, actually, was done experimentally, exactly to whom, and whether there was informed consent of any kind.

However, the paper appears to address possible future arm amputee needs.

Given that, it is interesting to see just how little actual prosthetic hands move in a way as it has been shown and tested here. The image shows interesting abduction and adduction measurements of this new flexible silicone, but that hardly makes sense for a prosthetic hand.

Critical issues

• Neural implants. Nanoparticles such as Ceria or Zirconia are used to suppress ROS (reactive oxygen species) enrichment which may be neurotoxic depending on specifics. The nanoparticles themselves are also potentially harmful; cells definitely react to such nanoparticles, and it is not totally clear how to avoid dissemination throughout the body. Yet, neural implants for direct signal hook-up seems to be crucial. Not just epimysial electrodes that are used for motor control. Then, ROS enrichment apparently can cause massive inflammatory responses and that can cause the death of nerve cells and damage the peripheral nervous system.
• Actual mapping. The very high density of nerves in a human hand correlates with the fact that a large area of our cortical brain maps to the hand, both for motor output and sensor input. That link is cut when the hand and arm part is amputated. Other brain areas are mostly not available, as they should be used for other stuff. Very high sensor density is certainly great, but before their output is fed into a person, why not coalesce them in some way.
• Grip and stretch. The better the grip, the less slip, the greater the friction and thus stretch forces. The worse the grip, the less useful the prosthetic hand as such. So really, before anything else, we need a glove that does survive simple things like carrying furniture or car washing before stretch forces actually become an issue.

Outlook

Where I do see an application for this type of sensor equipped silicone material is for control liners. So far, liners are mostly used for suspension (only). However, control increasingly becomes an issue, and thus control liners will be a relevant issue in the near future. Prosthetic hands that theoretically allow for single digit control are available, all that is missing is an easy to train control system. And liners can be more comfortable than they currently are.

That silicone product could be used to generate very dense and precise surface measurements of muscle contraction and displacement, as well as other variables, in order to extract actual grip patterns if not finger movements. That, in conjunction with surface EMG, is where I would assume a rather powerful future for this material, in context of below elbow prosthetic arm research.

@Article{jaemin2014,
  author = {Jaemin Kim and Mincheol Lee and Hyung Joon Shim and
     Roozbeh Ghaffari and Hye Rim Cho and Donghee Son and Yei
     Hwan Jung and Min Soh and Changsoon Choi and Sungmook Jung
     and Kon Chu and Daejong Jeon and Soon-Tae Lee and Ji Hoon Kim and Seung 
  Hong Choi and Taeghwan Hyeon and Dae-Hyeong Kim},
  title = {Stretchable silicon nanoribbon electronics for skin
     prosthesis},
  journal = {Nature Communications},
  volume = {5},
  pages = {},
  year = {2014/12/09/online},
  entrydate = {2014/12/10},
  abstract = {Sensory receptors in human skin transmit a wealth of
     tactile and thermal signals from external environments to the
     brain. Despite advances in our understanding of mechano- and
     thermosensation, replication of these unique sensory
     characteristics in artificial skin and prosthetics remains
     challenging. Recent efforts to develop smart prosthetics, which
     exploit rigid and/or semi-flexible pressure, strain and
     temperature sensors, provide promising routes for sensor-laden
     bionic systems, but with limited stretchability, detection range
     and spatio-temporal resolution. Here we demonstrate smart
     prosthetic skin instrumented with ultrathin, single crystalline
     silicon nanoribbon strain, pressure and temperature sensor arrays
     as well as associated humidity sensors, electroresistive heaters
     and stretchable multi-electrode arrays for nerve stimulation. This
     collection of stretchable sensors and actuators facilitate highly
     localized mechanical and thermal skin-like perception in response
     to external stimuli, thus providing unique opportunities for
     emerging classes of prostheses and peripheral nervous system
     interface technologies.},
}