He thinks an AI would probably have to build a quantum AI. He explained how a quantum AI would have completely different reasoning, and that having quantum AIs could bring about a "Quantum Singularity" -- the likes of which we cannot even fathom.
This site has covered quantum computers and now quantum computers can enhance machine learning.
Dwave has 128 Qubit superconducting chips now but there qubits are "dirty qubits". Their chips could hold tens of thousands of qubits if they were full. Millions of qubits that are more capable seems conservative by 2025. Full brain emulation AI systems will be developing at the same time as early quantum computers.
The Avogadro constant is the number of "elementary entities" (usually atoms or molecules) in one mole. The number of atoms in exactly 12 grams of carbon-12. It was originally called Avogadro's number. 6.022 X 10^23.
Early singularity scale quantum AI seems like it would be getting to mole quantitites of qubits along with the billions of times more processing power than the human brain.
Space Technology from Mundane to Amazing
A 200 Kilowatt Vasimr plasma rocket has been tested and could be raised to the megawatt level by 2015 and combined with small nuclear fission reactors to enable 39 day trips to Mars. The plasma rocket does not reduce the cost of launch to orbit (other than reducing the amount of fuel needed to brought up for space missions). This seems like an a certain level of capability that should very conservatively be in hand by 2020.
The science of nuclear fusion power and rockets is developing quickly. There is not much doubt that nuclear fusion power and rockets will be developed, but there most believe it may take decades.
Nuclear fusion can enable spaceships that can achieve 20% of the speed of light and inexpensive launch from the ground to orbit. The Winterberg deuterium fusion rocket design could have exhaust that is 6.3% of the speed of light. A multi-stage fusion rocket could achieve 20% of the speed of light with exhaust at that speed. There are various ways that the fusion capability which could be available by 2020 could become advanced space propulsion by 2030.
There has been a renaissance of funded nuclear fusion research
Mastering the mach effect (which relates to general relativity and possibly quantum gravity) would enable even more amazing capabilities.
An interview with Paul March on using the Mach Effect for propulsion
One-G constant acceleration and deceleration space drive would mean Earth-to-moon in 4 hours, Earth to Mars in 2-5 days, Earth to Saturn in 8-9 days. Advanced nuclear fusion could achieve performance close to this range as well by using microexplosions (external pulse propulsion variants of project Orion).
With fusion microexplosions and/or possible advanced mach effect multi-g acceleration is possible up to 100-Gs. The level of acceleration for a manned vehicle would be limited by what the occupants could endure. However, if the humans were enhanced to be able to withstand higher acceleration and were encased in a breathable fluid then 100-G acceleration could be possible. 100-G constant acceleration would mean Earth to Mars in about ten hours.
the Mach Effect for propulsion is proportional to the applied vxB Magnetic-field
The CUBE of the applied Cap Voltage
The CUBE of the MLT Operating Frequency
The SQUARE of the Cap dielectric constant
The thickness of the Cap Dielectric
Proportional to the total active Dielectric Mass
But Inversely Proportional to the Cap Density
So if material science can develop higher performing dielectrics and the Mach Effect can be leveraged as Woodward and Paul March have theorized then super high acceleration to near light speed could be achieved.
There is also the possibility of controlled wormhole formation for faster than light communication and travel.
Advanced nuclear fusion for power and propulsion and fully leveraging the mach effect would not require greater than human intelligence artificial general intelligence or quantum computers. These are capabilities that we can choose to develop now and put more resources and effort into developing.
If we do develop those capabilities and molecular nanotechnology then what would be the next level of technology that could come singularity enhanced humans or AGI ?
Perhaps developing and mastering femtotechnology ?
A 14 page pdf has been released by Professor Bolonkin Femtotechnology: Nuclear Matter with Fantastic Properties A lot of the paper is discussing what amazing things would be possible if it could be done. The key part is what is trying to be accomplished and the beginnings of how. We are talking about customized atomic nucleus strings and other shapes. The power of what would be possible is huge. But first what most would consider that impossible first step.
The form of matter containing and subsuming all the atom’s particles [from nucleons (neutrons, protons), electrons and other nuclear particles] into the nucleus is named degenerate matter. Degenerate matter found in white dwarfs, neutron stars and black holes. Conventionally this matter in such large astronomical objects has a high temperature (as independent particles!) and a high gravity adding a forcing, confining pressure in a very massive celestial objects. In nature, degenerate matter exists stably (as a big lump) to our knowledge only in large astronomical masses (include their surface where gravitation pressure is zero) and into big nuclei of conventional matter. Our purpose is to design artificial small masses of synthetic degenerate matter in form of an extremely thin strong thread (fiber, filament, string), round bar (rod), tube, net (dense or non dense weave and mesh size) which can exist at Earth-normal temperatures and pressures. Note that such stabilized degenerate matter in small amounts does not exist in Nature as far as we know. Therefore I have named this matter AB-Matter.