RobbieAllen.com > Blog

I attended the Council for Entrepreneurial Development’s big yearly conference for investors and entrepreneurs called Venture ‘06. The conference is an opportunity for early, mid, and late stage startups in the Southeast to present their business to a crowd full of venture capitalists in the hopes of securing additional funding. There seemed to be a direct correlation between the stage of the company and the quality of the presentation. Many of the early stage companies came off less than polished. Some of the later stage companies had compelling stories.

Noted inventor and futurist Ray Kurzweil did the keynote for the event. I’ve heard a lot about Kurzweil and his theories, but this was my first chance to see him live. He makes a pretty compelling case for where technology (and humanity for that matter) is headed. I took copious notes which are reproduced below. Perhaps my favorite quite was something to this effect: “In 15 years we’ll add 1 year for every year. So if you can hang in there, our life expectancy will move ahead of us.”

The slides from his talk are available at: www.kurzweilAI.net/pps/Venture2006/ and you can listen to the complete keynote courtesy of the CED Blog.

Notes:

- Decided he would be an inventor when he was 5
- Key to investing is timing
- 95% of projects he looks at as an investor fail because timing is off
- Hard to predict future of specific projects, but when it comes to predicting the impact of information technology (e.g., cost of MIPS, bandwidth increase, etc.) you can predict fairly accurate
- Info Tech is extremely chaotic, but the overall impact is predictable
- He gave a demonstration of new reading device for the blind
- It takes a picture of a page, then reads it back via synthetic speech
- Allows blind people to read labels on clothes, flyers, etc.
- Fits in a shirt pocket
- 10,000 times smaller than the first model 10 years ago
- IT isn’t the only field that is predictable: biology as well
- Fat insulin receptor drug will be on market in 5-8 years
- Fat insulin was useful when calories were scarce in the hunting/gather era
- Methods to turn off/on certain genes or add new genes
- Ability to turn off enzymes
- Starting to understand the brain - which is fundamentally an information process
- Key point is that not only is IT is taking over, IT power is doubling every year
- 25 yrs will multiple IT power by a billion and shrinking by 100,000 in 25 yrs
- Imagine what we have today, multiplied by these numbers
- Few examples of how pervasive these trends are:
- Rate of technology paradigm shift rate is doubling every decade
- Telephone took 50 years to be adopted
- Cellphone, web, social network took just a few years
- Evolution inherently accelerates exponentially
- The next stage goes more quickly than the previous
- Cambrian explosion went 100 times faster than before
- Homosapians took hundreds of thousands of years (almost a blink of an eye)
- Now we use the latest evolution of technology to improve technology
- computer-assisted software to design computers
- Been a very clear acceleration in technology improvement

- 1990 the Genome project was announced and skeptics said it would take a long time to decode the human genome. Half way through the project only 1% of the genome was matched, but then the rest was decoded very quickly.

- It is true that Moore’s Law is going to stop around 2020 when the processor features get too small (a few atoms) but then we’ll move on to the 6th paradigm of computing

- Showed lots of graphs of fairly smooth curves of the history of transistor price,
RAM, microprocessor clock speed and cost, magnetic data storage, internet hosts,
etc.

- We more than double our consumption of computing every year

- Biotech revolution: sharp contrast to the way biology used to work
- Early stages of being able to reprogram biology

- Eventually replace animal testing with computer simulations

- Looking at logarithmic graphs show clear trends with exponential growth whereas linear graphs make it seem the trends come out of nowhere

- While biology is intricate it is suboptimal in many ways

- Roboitc red blood cells are much more efficient than human red blood cells. Will allow you to run at Olympic race speed for 15 minutes or stay at the bottom of the pool for 4 hours
- If you consider the 25 year trend line, these progressions are quite possible

- Gave a lot of examples of AI-based programs that are widely in use today (Narrow AI)

- Knowledge about the brain is doubling every year.
- Will have the full algorithms of the brain by the 2020s
- While the brain as billions of connections, it isn’t that complex because the genome isn’t that complex
- We’ve reversed engineering the design of the brain and are able to understand it
- Compared the brain to a mandlebrot set
- The brain is essentially a probabilistic recursive fractal

- Ecommerce revenues has been growing exponentially and not blip to do the dotcom bomb
- 2000-2003 was a Wall Street phenomenon: exponential growth does not mean instantaneous growth :-)

- Translating language will be a common feature of cell phones
- Showed a demo where we a person spoke english and it was automatically translated to different languages

- 2010:
- Computers disappear: they are getting smaller and smaller
- Images written directly to our retinas
- High bandwidth connection to the Internet at all times
- Electronics so tiny it’s embedded in the environment: clothing/eyeglasses
- Effective language technologies
- Augmented real reality
- Interaction with virtual personalities as a primary interface

2029:
- 30 doublings by then
- $1000 of computation = 1000 times the human brain
- reverse engineered the brain
- nanobots provide:
- neural implants that are nonevasive surgery-free, distributed to millions or billions of points in the brain
- full immersion virtual reality, design of new bodies will be an art form
- expansion of human intelligence, multiple our 100 trillion connections many fold

- Emotion is the cutting edge of intelligence, but thinks by the 2020s we’ll have
loving and caring computers

Avg Life Exptenacy:
- Cro Magnon: 18
- Ancient Egypt: 25
- 1400 Europe: 30
- 1800 Europe and US: 37
- 1900 US: 48
- 2002 US: 78
This is going to go in hyperdrive as we engineer around disases
In 15 years we’ll add 1 year for every year
So if you can hang in there, our life expectancy will move ahead of us

We are the species that goes beyond our limitations