Open the #pod bay doors, @HAL
Despite over promising the likes of HAL 9000 the Artificial Intelligence (AI) community has been making steady progress. Indeed, the famous Watson Jeopardy Experiment was a great demonstration of the coming era of "smart systems." Other examples are Apple's Siri, and smart search engines (including Google, which seems to be getting smarter about its search results each year.)
All of these efforts have several things in common; AI based software, piles of data, and racks of commodity hardware. Popular conversations include terms like business intelligence, knowledge discovery, Big Data, Hadoop, and other new buzz words. Is this yet another fad being oversold by the marketing types or is this a game changing set of technologies that will shape how we interact with almost everything we touch?
To help answer this and other questions, consider a few recent events.
- Lenovo buys IBM x86 server business for $2.3 billion
- IBM to invest $1 billion to create new business unit for Watson
- Google has agreed to acquire privately held artificial intelligence company DeepMind Technologies Ltd.
- Ray Kuzweil's Job is to Beat IBM's Watson at Natural Language Search.
HPC AI Is Now Big BusinessIn the space of a few weeks, roughly $1.4 Billion has gone into AI efforts at IBM and Google. This sum includes the $1 Billion pledged to Watson development and the rumored $400 million Google paid for DeepMind. IBM of course is flush with cash from the Lenovo sale ($2 billion cash and $300 million in Lenovo stock).
The sale of IBM x86 server division is intriguing. This sale includes System x, BladeCenter and Flex System blade servers and switches, x86-based Flex integrated systems, NeXtScale and iDataPlex servers and associated software, blade networking and maintenance operations. In other words, they sold the farm. One would think IBM has a strategy behind this sale because the x86 server business has been doing well of late. [UPDATE: Well not so great actually.] One might infer IBM considers Watson in the Cloud as the way forward.
In terms of Google, they want to deliver a "better Watson" though the acquisition of DeepMind and similar technologies. Of course like IBM, Google has a bunch of "smart people" working on the "smart cloud." The competition should be interesting to watch. That is from what we can see sitting outside looking in.
Finally, tools like Hadoop have been making it easy for virtually any company to get into the Big Data game, which after counting words, is often about business intelligence. The advent of Apache Hadoop YARN is also introducing more capability to the Hadoop cluster. (i.e. Hadoop is moving beyond Map-Reduce.)
Behind The CurtainAnd now for the scary part. The ability to intelligently analyze large (or small) amounts of data can be a great tool. Intelligent tools can help solve many pressing problems in science, engineering, business, education, government, and many other areas. These tools are very powerful and potent. If used in a good way they can raise the tide for every one. If on the other hand, they are used to invade personal privacy or broach surveillance laws, things may get problematic for those whose lives create the mountains of data fed into the smart clusters. Social media, it seems, is fertile ground for smart analysis. Inferring things about you from your twitter feed (or anything else you write for that matter) or what store your phone happens to visit are a bit eerie. A recent discussion of some new capabilities of IBM's Watson technology can be found by reading the recent Datanami article; Why IBM Knows You Better Than You Think (Watch the video to understand what these tools can do with social media data.) The final paragraph in the article is telling:
"The technology can do all of this without having official access to customers' Twitter accounts. There are enough clues left on publicly visible social media accounts for the technology to identify the real-life person with a high degree of confidence. Those who are averse to having their social media data mined in this way can change their social media privacy settings. IBM says it is also working to institute a more global opt-out feature."
Think about that for moment. And remember, these systems are just getting started and not limited to IBM or Google. Billions of dollars are being invested new intelligent designs and opting out of the world is not an option for some people. Finally, for the chronologically gifted readers, let's hope someone still remembers to say,"Smile, you are on Candid Camera."