Wednesday, March 30, 2016

Vision vs. Ultron: The Future of Artificial Intelligence

The purpose of IS is to be able to use technology to better perform tasks and integrate technology to make a process run more smoothly, quickly and accurately.  Typically this is done by human analysis and research in order to find ways to better use technology. However, what if there was a way to circumvent the need for human inter-mediation?   This is where Microsoft's, as well as other large name companies such as Facebook and Google, are attempting to break new ground. The use of Artificial Intelligence is becoming the status quo.  Companies are encouraging their consumers to interact with the AI in order to things like get help with a product, order something offline, or other menial tasks like schedule an appointment.  However, the past of AI is muddled with mistakes. In recent news, the failed release of the AI "Tay" remarks some of the caution that most people have against trusting AI.  Within 24 hours of its release, the bot was taught to praise Hitler, claim Bush orchestrated 9/11 and show a strange affinity for illicit drug use. In theory, AI being taught to handle user input is a fantastic venture.  However, the learning for the AI must take place in a safe environment where they cannot be so easily corrupted. Once the AI has learned some basic do's and don'ts its potential for success is limitless. Amazon's "Echo" utilizes AI to help make suggestions to their user and to process commands.  Much of the success for the AI seems to depend on how it is implemented and certain rules/knowledge embedded in its creation.
        The article does a good job of analyzing the failures of AI, but it only slightly mentions the successes.  AI has a very long and arduous journey to even come close to being as widely accessible, and successful, as depicted in some science fiction. It also does not seem to highlight that ventures into the world of AI have been as early as the late 90's and early 2000's. AI has come a long way since then, but still has quite the trek remaining.
        The article also could do a better job representing the interactions of AI today.  It does mention complex and rare examples like "Echo", but neglects even the simpler forms of AI like the bots in computer games. These are much more common place, but not as typically thought of as meaningful forms of AI.
        AI has a future in the world. Some people are ready for it, others are very hesitant because of examples like "Tay". Hopefully examples like "Echo" will persuade the masses that AI is meant to be for our betterment. This article is a decent means for introducing the reader to a world with which they are not familiar.

http://www.bloomberg.com/features/2016-microsoft-future-ai-chatbots/

4 comments:

  1. Since people historically have been very hesitant to accept artificial intelligence, I wonder if it will ever hold a serious role in our society. Many people fear AI because they think robots will outsmart the humans that created them. As you mentioned, Echo and computer gaming bots have already assimilated into our everyday lives as well as robotic medical equipment, self driving cars, and different security measures.

    ReplyDelete
  2. While AI certainly is viewed negatively by many people, I feel that these fears will dissipate with familiarity. Many people's only experience with AI is what they see in movies and video games, or the failed experiments that they read about, and so they have doubts, as AI is often portrayed negatively in these mediums. However, as useful AI become more commonplace, and the tech starts to help people better society or achieve tasks more efficiently, I think that people will trust AI, and it will become as commonplace as computers and mobile phones are currently. Examples of Ai such as "Echo" will certainly start to change public sentiments about AI, and the naysayers will likely soon be in the minority.

    ReplyDelete
  3. I personally have a negative opinion on AI as it lacks a sense of human touch and emotion. I think as a society, we have a hard time connecting with AI as it seems advanced and unnecessary for our time. While it may have professional uses, I can not understand why major tech companies are eager to create an AI system to talk to their customers or tech users as a whole. Also, I’m not sure that a mainstream market for AI communication will be created anytime soon. This to me is displayed by the references to Hitler, Bush doing 9/11 etc. done by “Tay.” I’m pretty confident that it wasn’t the majority of internet users coaching Tay into saying these things. Instead, it was a very small group of internet trolls who are interested in this technology and using their time to demonstrate flaws within it. Personally, I think it will be a rather long time before society becomes more accepting and interested in AI. Until then, I think AI will remain a project of tech companies and a toy for certain niche markets.

    ReplyDelete
  4. The concept of Artificial intelligence (AI) has struck curiosity, but also fear, into the people of today’s society. The image depicting AI by many science fiction films and comics is of man-made technology outsmarting humanity, and taking control of the world. When analyzing this blog post, and the article attached to it, one can see that it is easier to depict a failure than to portray a success.
    The AI technology mentioned in this post, “Tay”, shows its learning and assimilation to inhumane concepts and thoughts. What some may not realize is how, and what is being taught to the technology, and what it is exposed to. I agree with this post, in that AI cannot be introduced to an unstable environment unless experts know that it is capable to see the difference between right and wrong. In order for the public to become less averse to the concept of AI, technology must prove itself time and time, again.
    The one thing that I would like to know more about is, how far will scientific technology go, in terms of AI’s abilities? Will the increase in advances cease once we feel that technology is too advanced? For example, as it seems in this post, AI has increased its ability to do things that humans can do, and eventually do those things without error. Will that lead to a decrease in jobs for humans in the future? This could be a potential concern that is worth keeping into account. Overall, this post was very intriguing and insightful, and I am excited to see what AI has to offer to society in the future.

    ReplyDelete

Note: Only a member of this blog may post a comment.