Traditional computer chips are getting close to physical limits. At the same time, energy use from AI data centers is projected to double by the end of the decade. This creates questions about sustainability and design.
Scientists are exploring neuromorphic computing, hardware that works more like the brain. The brain performs complex tasks using about 20 watts, roughly the power of an old light bulb. Researchers say today’s computer architecture uses much more energy.
Teams are building devices that mimic synapses and put memory and processing in the same place. That could cut the energy used to move data between parts and improve efficiency.
Difficult words
- physical — about the material world and natural limits
- sustainability — ability to continue without harming environment
- neuromorphic — made to work like the human brain
- synapse — small gap where brain cells send signalssynapses
- architecture — basic design and structure of a computer
- efficiency — use of little energy to do work
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- Do you think computers that work like the brain are a good idea? Why or why not?
- Can you name one thing at home that uses about 20 watts?
- Why is energy use an important question for AI data centers?
Related articles
App and text coaching help university students' mental health
A multi-university trial tested a smartphone app plus personal text coaching to reduce depression, anxiety and eating disorders. More than 6,200 students took part and reported fewer symptoms up to two years after starting the program.
Brain differences in WTC responders with PTSD
New imaging research of World Trade Center responders finds measurable brain structure differences linked to long-term PTSD. Researchers used gray-white contrast (GWC) MRI and other markers to distinguish responders with and without PTSD.
