I started reading The Singularity is Near today – this is Ray Kurzweil’s new book. He gave a talk on the Microsoft campus last week but I was unable to attend. I’ll have track down the webcast from that event because I’d really like to hear him speak.

The thing I like about Kurzweil is that fact that he’s not afraid to think big. He has this relentless optimism about the power of technology to fundamentally transform human existence. That’s essentially the theme of Singularity – that humanity’s technological evolution is advancing faster than its biological counterpart, and there will eventually come a time when we stop thinking about ourselves as purely biological organisms. If you believe Ray, that time actually isn’t too far off.

I keep thinking back to a book I read a while ago called Natural Born Cyborgs. The title makes it sound more science-fictionish than it really is; it’s written by Andy Clark who has a pretty impressive set of credentials to back him up. One point he made in that book is that the human brain is a tremendously flexible and adaptive system – it can transparently take advantage of externally available tools. For example, how many times have you asked someone if they know the time only to have them respond “yes” and then immediately look at their watch? It’s a simple example of how knowing how to retrieve information is often just as good as actually knowing it. Intelligence is not a function of what you know; rather, it’s a function of what information you can reliably retrieve within a reasonably small period of time.

Over the last ten years, the amount of readily accessible information has increased at a staggering rate. There has been a corresponding increase in mobile technology and connectivity. The net effect is that we’re spending less and less time disconnected from this vast corpus of knowledge, and our brains (well, maybe it’s just me…) are starting to depend on it being there. Rather than remembering facts, I find myself remembering pointers to facts – search strings I can punch into an engine to retrieve the facts I need when I need them. As long as I’m connected, the cost of dereferencing these pointers is relatively low and the time/space tradeoff is quite favorable.

Today, the data pipe between the L2 cache in our biological memories and the data stored in non-biological memory (readily accessible persisted information) is pretty slow and unreliable, which limits the degree of integration between our internal cognitive processes and external computational aids. However, this is rapidly changing – the penetration of wireless networks and the corresponding increases in the compute power of mobile devices means that we are spending less and less time disconnected from our non-biological memories. As such, we will become increasingly reliant on access to those memories and they will be more fundamentally integrated into our routine cognition[1]. Our interfaces to non-biological memory are also evolving at a rapid rate; they are moving away from the visual/textual paradigm and shifting to auditory/speech[2]. Tons of research is going on in the field of natural language interfaces, and I don’t think it will be too long before we start interacting with computer systems in much the same way we interact with people. The ultimate extension of this interaction is some form of direct neural human/computer interface – a concept which I find both extremely compelling and more than a bit unsettling. I think we’re still a ways off from this, but the foundations are being laid today. We already have things like cochlear implants that plug into the brain’s existing neural structure and supply it with artificially created sensory input. There’s a large jump between interfacing at the syntactic level (sensory input) and the semantic level (deep integration into higher-order brain functions such as language and memory) but I don’t think such advances are fundamentally impossible. It’s really just a matter of time.

That, I guess, is the foundation of Kurzweil’s optimistic view about the future – his unrelenting belief in a sustained exponential increase in the power of technology over time. I find this optimism refreshing. I find his belief in the power of technology inspiring, and I find his vision of the future just scary enough to be pretty exciting. I’m glad there are smart people out there who are willing to stand up and say some pretty far-out things. The first step in making great things real is to believe that great things are possible.

[1] Case in point: since I got my Smartphone, I have basically stopped explicitly remembering where my meetings are scheduled. There’s no need; my phone is functionally the part of my brain responsible for remembering where my meetings are.)

[2] My car has a voice-enabled navigation system. I can literally ask it to “find the nearest Japanese restaurant” and it will select one and tell me how to get there. This is very first-generation technology; I’m waiting for the day when I can ask it to “find the nearest good Japanese restaurant” and it will return the results based on my own personal preferences and the recommendations of trusted members of my digital social network.