My Predictions for 2027

Shanghai at night

On this site I’m planning to do more than my fair share of criticizing the predictions and visions of the future of others so I figured to I should put my own neck on the line regarding the road to the singularity and throw out some predictions for a 2027, a decade from now. Fingers crossed this site is still online in 2027.

Predictions for 2027

  • 5 Year survival rates for cancer have doubled

    By 2027 cheap genetic sequencing and synthesis will enable rapidly iterated personal therapeutics against cancer. Drugs will be tailored to your own cancer’s unique genetic code, rather than the one-size-fits-all approach of today’s treatments. As a cancer patient in 2027 you will routinely have your cancer sequenced to identify new mutations, these mutations today are often fatal as the cancer can adapt and resist the drugs used against it, but with rapidly iterated personal therapeutics the drugs could change as fast as the deadly mutations.

  • The first SENS therapies against heart disease and arterial sclerosis are available

    Research funded by the SENS Foundation has demonstrated the ability to protect cells against 7-ketocholesterol toxicity, a major contributor to heart disease and arterial sclerosis. Heart disease and arterial sclerosis are the two biggest killers in the developed world with over 600,000 deaths annually in the US alone. The SENS research team used a modified microbial enzyme that can breakdown and eliminate 7-ketocholesterol inside the lysosome which prevents the lysosomes from turning into foam cells. It’s these foam cells that build up inside the arteries and can eventually burst leading to heart attacks and strokes. Dr. Aubrey de Grey,  founder of the SENS Foundation has said human clinical trials could be possible in the 2020s given adequate resources for research.

  • The majority of new cars are fully electric

    Global electric car production is currently growing at more than 50% per year, far faster than the growth of internal combustion engine cars. Multiple car manufacturers are projecting they’ll reach a million-plus unit electric cars produced annually by 2020. China’s electric car production is currently growing the fastest and their government wants 11% of all new cars to be electric by 2020, if they can maintain their current growth having a majority of new cars electric by 2027 is actually conservative. It’s a similar story for India, the Indian government recently announced they want all new cars to be electric by 2030.

  • China is the world’s leader in science and technology research

    Unless the United States radically changes it’s research and development funding plans -an increasingly unlikely prospect given the current administration- China will easily overtake the US in both basic science funding and in the number of new papers published annually by 2020 and by 2027 China will be dominate.

  • Self-driving trucks and taxis are commonplace

    Almost every major auto manufacturer is currently working on autonomous vehicles as are some of the biggest tech companies such as Google, Uber, and Apple. These companies have already driven millions of miles on public roads and they’ve done so at a far safer accident rate than human drivers could have achieved. The economic forcing function for development autonomous vehicles is enormous, trillions of dollars could be saved by eliminating human drivers and enabling vehicles to operate around the clock, a self-driving truck isn’t going to require any government mandated rest time. I’ll bet you and I will have hailed a self-driving taxi before 2027.

  • Atomically precise manufacturing will not yet have been achieved

    My own belief is that atomically precise manufacturing will require human-level artificial general intelligence in order to be developed. I suspect atomically precise manufacturing could be developed by human engineers and scientists but the scale of the project that would be required is far beyond anything that any government is likely to commit to, we’re talking perhaps thousands of engineers and scientists, which is on the order of the Apollo program. On the other hand, if you have human-level artificial scientists and engineers at your disposal the project becomes a far cheaper one, one that a government or corporation could reasonably fund. I expect atomically precise manufacturing to be developed with the aid of AI engineers and AI scientists sometime in the 2030s.

  • Human-level artificial general intelligence will not yet have been achieved

    Vernor Vinge and Ray Kurzweil among others believe human-level artificial general intelligence (AGI) will more likely than not be achieved in the 2020s. I’m somewhat more skeptical, I agree with them that the prerequisite computational power will be available in the next decade but I bet the algorithms required to really bake the AGI cake will take some years longer.

    My own guess is that we will see extremely impressive AGI in the 2020s and in some domains it will undoubtedly be superhuman but there will still be kinks to be ironed out. I suspect you will still be able to tell after significant interaction with an AGI of 2027 that you are dealing with a non-biological intelligence, in the early 2030s that will no longer be possible and AGI will therefore have passed the Turing Test.

  • A full human genome sequence will cost less than a dollar

    In 2016 the cost to sequence a full human genome fell below $1000 for the first time. Since the start of the Human Genome Project in 1990 the cost to sequence has fallen significantly faster than the speed of Moore’s Law.

Cost per full genome sequence by year, graph courtesy of the US National Institutes of Health
Cost per full genome sequence by year, graph courtesy of the US National Institutes of Health

By 2027 the cost to sequence a full human genome should be well below a dollar, and eventually it will be come so cheap and easy to do that your toilet will be able to sequence your

  • A full human genome will be synthesized

    To sequence the human genome it took 13 years, a worldwide scientific effort and almost three billion dollars in funding. Scientist are now looking to do the same again, but this time instead of reading the human genome they want to write it. Scientists, including Prof. George Church of Harvard, announced the Human Genome Project-Write last year which they believe could produce a synthesized human genome in ten years.

    As of 2017 the largest synthesized genome is that of bacteria with a few hundred thousand base pairs of DNA to synthesize, comparing that to the 3 billion base pairs in the human genome means serious -but not unprecedented- exponential growth is required for the Human Genome Project-Write to finish within a decade.

  • Batteries will be below $75 per kWh making new electric cars far cheaper than internal combustion engine cars

Tesla Motors Model S base, seen at Santana Row in San Jose, California
Tesla Motors Model S chassis and battery, seen at Santana Row in San Jose, California – Image courtesy of Oleg Alexandrov, Wikimedia Commons

The point at which new electric cars reach price parity with internal combustion engine cars without any subsidies is estimated to be when batteries can be produced for $100 per kWh. Tesla claims it has been capable of producing batteries at its new Gigafactory for less than $190/kWh since 2016, which is four years earlier than McKinsey’s energy analysts had predicted it would happen.

Competition in the battery world is heating up and with it the pressure to reduce the price of batteries, so by 2027 I don’t believe it is at all outrageous to predict that production costs will be below $75 per kWh, which should make anyone who relies solely on the oil industry and internal combustion engine cars for their business shudder.



Facebook Comments