Blogs

Barriers to entry for early adopters of IBM Watson

Director of Interactive Sales, WNEP-TV / WNEP.com

There have always existed significant barriers to entry for early adopters of any new technology. For example, all the miles of cable that needed to be run in order for the world to take advantage of the miracle of electricity. Now consider the computer: devices that were capable of processing only the most rudimentary of algorithms used to fill rooms, and were programmed using cryptic punch cards. Now the world has been introduced to IBM Watson, the first and only (as of the time of this writing) commercially available artificial cognitive intelligence.

Chasing Watson

I was immediately fascinated by the possibilities this technology that would one day change the world. Watching Watson best the greatest Jeopardy champions sent chills down my spine: we had witnessed a machine that was not only able to (almost) flawlessly understand the spoken, natural language of the show, it was also able to provide incredibly accurate contextually appropriate answers in sub three second time. The power of this technology was unmistakable, but I was left thinking that this wonderful new intelligence would only be relegated to the most advanced institutions of higher education and the deepest and darkest corridors of government research. Watson was cool, but unattainable.

Fast forward many months, and I began to hear rumblings in the tech community that Watson might not be that far off. First it was a TechCrunch article about the impending launch of Watson Analytics, then some tantalizing social media activity providing snippets of information—I was officially back on the hunt for Watson.

Burning questions

barrier to entry Watson.jpgEven though it appeared as though Watson tech might be slightly more available, many questions remained:

What kind of limitations would be imposed on a potential consumer-facing version of this system?

Would this only be appropriate for use in very specific business verticals?

What kind of development deep dive would be necessary to properly implement Watson?

How many millions would this cost?!

Insight on Watson

IBM Insight was truly an eye-opening experience that has not only answered my questions, but has created a full-on Watson devotee without my even having laid a finger on a development console! You see, I've learned that humans produce data at a staggering rate. We will create more data this year than in all previous years combined. Unfortunately, however, greater than 80 percent of this data is unstructured. That means that less than 20 percent of all data is stored in such a way as to make it useful. This unstructured dark data will only grow over time and become more of a drain on our technological resources due simply to the fact that we must pay to store it, but are almost completely incapable of using it in any meaningful fashion.

Enter IBM Watson. The cognitive intelligence of Watson technology provides companies with the unprecedented opportunity to ingest and understand dark, and structured, data in its entirety. Not only Will Watson ingest this data, but it will analyze it contextually and provide a meaningful way to interact with it.

Take, for example, the work being done at Sloane Kettering Memorial Hospital. Through its partnership with IBM, Sloane Kettering has fed millions of documents of the latest in cancer research into Watson and then reinforced it with the knowledge and experience of its world class staff. By doing this, Sloane Kettering has created a cognitively intelligent Cancer Specialist who may act as a trusted advisor in the treatment of an unlimited variation of patient types. Now take this data, and replicate it to health facilities across the globe. This immediately provides the virtual equivalent of a world-class Cancer Treatment Physician to institutions who need it most.

Accessible, applicable, available

Beyond the capabilities of this new Watson technology, the most impressive part of this entire application ecosystem is that it is not unattainable. You could sign into IBM Bluemix, sign up for a 30 day free trial and be up and running with an instance of Watson Technology today! That account provides access to the following:

•IBM Watson Development API Access

•Bluemix Development Platform Access

•Training resources and videos

•The ability to write Watson applications in several languages

If you aren't interested in necessarily developing a new application from the start, then you may be interested in taking advantage of any of IBM's already developed apps:

•Watson Oncology Advisor

•Watson Wealth Management

•Chef Watson (no it doesn't go so far as to actually cook for you, yet!)

If you are an enterprise user who is interested in a more robust Watson offering, you have even more options in terms of physical Watson implementations in your private datacenter, or private managed instances in the cloud.

I could go on about Watson for pages, but I'm going to conclude by answering the questions I had posed above:

What kind of limitations would be imposed on a potential consumer faced version of this system? There are no limitations imposed on this consumer-facing version of Watson.

Would this only be appropriate for use in very specific business verticals? This is an extremely open system that could be, and should be, used by any business vertical.

What kind of development deep dive would be necessary to properly implement Watson? Watson actually takes the impossible challenge of cognitively and contextually cataloging an organizations data out of the hands of developers. This will ultimately make your life, and your business, better.

And lastly, how many millions would this cost!? Regardless of the low actual cost of startup, I'll answer this question with a question: Can you afford not to use Watson?