The Speculist: The Other Singularities


Live to see it.

« What Changes? What Remains the Same? | Main | Somehow I missed this... »

The Other Singularities

[Electricity from solar power] is doubling now every two years. Doubling every two years means multiplying by 1,000 in 20 years. At that rate we'll meet 100 percent of our energy needs in 20 years.

- Ray Kurzweil

Suppose that solar power achieves and then surpasses "grid parity," meaning that it produces electrictity less expensively than conventional power sources. Once that happens, solar power would inevitably become the dominant energy source. Call it the Solar Singularity.

- Arnold Kling

Arnold Kling spends much of the rest of his article arguing that Kurzweil's solar optimism is misplaced. I disagree, but Kling's "solar singularity" idea is worth exploring.

Kurzweil's extrapolation on solar is simplistic. But it will also be right. It's not that Kurzweil thinks that Solar will improve (just) because we'll be able to write more solar cells in a smaller space like transistors on a computer chip. Instead, Kurzweil is describing a fundamental law of progress of which Moore's law is only the latest iteration.

Kurzweil (and others) have described the power of exponential doubling with a "Rice and the Chessboard" parable:

A king wanted to reward his faithful wiseman. The wiseman stated that all he wanted was a single grain of rice for the first square on a chessboard, 2 grains for the second square, and so on... doubling until all 64 squares were filled.

The King ordered that this "modest" request be honored - until he learned that his wiseman was really a wiseguy. 64 doublings would be all the rice ever harvested in the entire world.

Enter "Spock's Chessboard"

Exponential trends are powerful in ways that are hard to foresee even with something as simple as rice. But the doubling of information technology is much more surprising.

Moore's law (and its predessesors and successsors) have been progressing through an information technology chessboard. Every two years or so puts us on a new square with twice the computation per dollar spent. But that's not all. Because computation is the basis for the development of everything else, the computation board spawns other exponential chessboards. Solar technology gets its own board. Other energy technologies like fusion, biodiesel, and alcohol fuels are also progressing on their own chessboards.

Life extension gets a chessboard. In fact, Aubrey de Grey has already described a life extension singularity. It's his "bootstrapping to escape velocity" idea.

Each mini-board has its own singularity. Kling has described the solar singularity. A similar singularity could happen for fusion power. The age of personal medicine will arrive, I think, about the time that a $1,000 genome sequencing becomes possible. That will be a singularity for the genetics mini-board.

Along the way each mini-chessboard is fed by the main computation board, but they also sometimes feed other mini-chessboards or even feed the progress of the main computation board.

It's hard to overstate the challenge of forecasting the future of technology with so many interlocking variables. But with so many exponential processes working in our favor, I'm not going to bet against Kurzweil's optimism.


I wrote about a solar majority and a wind majority power grid last year based on an excellent 2006 SunPower Talk, and similar data:

Search for: "SunPower Talk - US To Be Majority Solar in 19 Years? ", as well as "Windmills US Majority Power in About 18 Years?"

Similarly "Hybrid Autos at 50% Market Share in 10 Years?" predicts a hybrid takeover of the US auto market in 2017. That's interesting to me, because if it happens, a follow on takeover of plug-in hybrids, which would vastly reduce liquid gas consumption, would be by 2027.

We'll probably stop importing gas before we pull out of Iraq under most current plans. Not that there's any relationship. :)

The Fusion Sigularity may be at hand:

WB-7 First Plasma


Intermittent power sources are not going to be of much use without efficient low cost storage.

Which currently has no road map. Unlike solar and wind.

M. Simon:

Batteries have their own mini-chessboard, so do ultra-capacitors.

It's hard to know yet which form of storage will dominate.

By the way, I don't think that we'll be 100% solar in 20 years. I doubt that even Kurzweil thinks that.

I do believe that solar could quickly become the dominant power source given sufficiently cheap cells and storage capability.

I wholly agree that solar is due for a huge burst, NPI, of exponential innovation.

According to a recent Future Blogger post by a solar industry buddy of mine, Erik Brolis, "locations such as Hawaii are already at 'grid-parity,' where the cost of purchasing electricity from the utility equals that of producing one’s own from an unsubsidized solar PV system. Most PV manufacturers anticipate grid-parity within 5 years in most places around the world. Even George W. Bush proclaimed the goal of grid-parity by 2015!"

That's the now and the near-future. Subsequently, advanced biological modeling + robust evolutionary computing + AI-ish thing + advances in all other fields will push the exponential solar trend.

The link to Eriks post is:

Market forces are likely to delay the Solar Singularity, like they have in the case of the electric car. I hope twenty years is possible, but there is likely to be interference.

Hydrogen, which has mistakenly been touted as an energy source, might well provide the needed medium for energy storage for solar.

By the way, I think 64 doublings of a single grain of rice =


Post a comment

(Comments are moderated, and sometimes they take a while to appear. Thanks for waiting.)

Be a Speculist

Share your thoughts on the future with more than


Speculist readers. Write to us at:

(More details here.)



Powered by
Movable Type 3.2