Moy Gulati
← Writing

Biocompute musing 4 (Biological computation) for General AI

2 min read aibiological computationhardware

Hardware is rather more important than software to achieve general AI!

The singularity in artificial intelligence (AI) is the point at which AI becomes capable of self-improvement and surpasses human intelligence. Achieving this point will require significant advancements in both hardware and software. However, it is argued that hardware-level developments are more crucial for achieving singularity than software-level developments.

  1. Computing Power: One of the key requirements for achieving singularity in AI is the ability to perform complex computations at a rapid pace. This requires a significant increase in computing power, which can only be achieved through advancements in hardware. This includes the development of new processors, memory devices, and interconnects that can perform complex computations at a faster rate.
  2. Neuromorphic Computing: Achieving singularity in AI will require the development of neuromorphic computing, a type of computing that mimics the function of the human brain. This requires the development of new hardware, such as neuromorphic chips, which can perform complex computations at a faster rate than traditional processors.
  3. Quantum Computing: Quantum computing has the potential to revolutionize AI by allowing for the manipulation of data at the atomic and subatomic levels. Achieving singularity in AI will likely require the development of new hardware, such as quantum computers, which can perform complex computations at a much faster rate than traditional computers.
  4. Energy Efficiency: Achieving singularity in AI will require the ability to perform complex computations at a rapid pace without consuming excessive amounts of energy. This requires the development of new hardware, such as low-power processors and energy-efficient memory devices, which can perform complex computations at a faster rate without consuming excessive amounts of energy.
  5. Scalability: Achieving singularity in AI will require the ability to scale up AI systems to handle large amounts of data. This requires the development of new hardware, such as high-performance storage devices and interconnects, which can handle large amounts of data at a faster rate.
  6. Robotics: Achieving singularity in AI will require the development of intelligent robots that can perform complex tasks. This requires the development of new hardware, such as robotic actuators and sensors, which can perform complex tasks at a faster rate.

I would also think, that we have seen the hardware shift happening to support better efficiencies, for example, consider a calculator - 400 years ago, we had an abacus (Hard rocks, beads), then we found a better way to solve the problem (Software) and moved on to calculators (plastic, silicon), now the same calculator is sitting in your apple watch using completely different chemistry and physics of the same base elements. My hypothesis is simple, we work on improving the software but after a point, it requires a paradigm shift that would warrant a hardware-level change (mostly a completely different way of solving the problem)