I attended the Singularity conference in San Jose last weekend to catch up on the latest and greatest advancements in AI. I had lived through and participated in the first wave of AI. Worked at the AI lab at MIT, conducted a technology transfer, created a product and a launched a company on the back of AI technology. AI had a big impact on my professional life and I largely ignored it during the internet boom years and the emergence of web 2.0. So what is this new found obsession with AI?
The setting of the conference was eerily familiar. During the first wave of AI optimism I attended an AI conference at the San Jose Convention Center. The Singularity conference was across the street at the Auditorium and Tech center. The first conference had at least a 1000 people attending the Singularity had perhaps 500. This in itself represented a change that turned out to be a positive and not a negative. During the Friday panels it became clear that people were more sober and realistic about what AI technology could achieve in the short run. Researches and business people had learned that predicting the economic viability and practical implementation of research was “hard”. It takes more time and money then you would like to turn a laboratory experiment into something useful and commercially viable.
This realistic approach was refreshing and encouraging because it did not negatively impact the enthusiasm people had for the science and technology. The panelists demonstrated maturity in their approach to commercializing technology and predicting timelines associated with transferring science into the commercial realm.
The closing speaker at Friday’s panel made a subtle but profound statement that was not only picked up by me but many others at the conference. He indicated that the Singularity may be realized incrementally over a relatively long period of time. These incremental changes may be so gradual that no one actually finds them amazing. If and when a Singularity does occur no one will be particularly shocked. Oh well machines are as smart as humans and perhaps even smarter let’s all move on.
The realization that the Singularity may take a very long time did not really phase people. I think the attendees respected the fact that the pursuit of the Singularity was even more important then the Singularity event. This pursuit would result in all kinds of innovation that would be valuable in and of themselves. An example, the average person is unaware of the fact that Google became successful as a company based on their use of fuzzy logic case based reasoning searching techniques. I suspect that there will be many more Google examples in the near future.
2 comments:
I agree that the attitude towards the applications of AI was realistic and humble.
However, the slow research in AI over the past few decades has indeed finally launched some very interesting projects which wouldn't have been possible a decade ago, of note are openCog(www.openCog.org)[preparing intelligent pets in secondlife],Kismet(http://www.ai.mit.edu/projects/humanoid-robotics-group/kismet/kismet.html)and IBM Blue Brain Project.
I don't fully agree with the closing note of Friday's speaker. His argument makes sense if technology was something linear. I think that as time proceeds not only will new applications of technologies become more readily available but at the same time the web of technology industry will become more complicated by faster discovery of new niche's and new technologies, a good example is rapid introduction of social applications in web, cell phone and bioinformatics industry(http://openwetware.org). I think people would definitely be in awe of that, and the rapid change and novelty wouldn't go unnoticed.
Good point perhaps the increasing rapid rate of change brought on by information technology will in fact amaze us. Certainly, the viral aspect of social networks and the rapid launch of new services and products on the web startles me.
Post a Comment