Scribd Logo
Upload

Welcome to Scribd!

  • AI Faults: Keywords:-AI AI-faults Risks of Ai Human Oracle

    Uploaded by

    International Journal of Innovative Science and Research Technology
    14 views3 pages
    The paper is written to the development of the AI-faults, commonly known as AI-hard. Now the question arises should we resolve the faults? Is the faultsresolving beneficial for humankind? The paper suggests some important AI-faults & as well as the topic if we resolve this faults what could be happened to world.

    Original Title

    AI-Faults

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    The paper is written to the development of the AI-faults, commonly known as AI-hard. Now the question arises should we resolve the faults? Is the faultsresolving beneficial for humankind? The paper suggests some important AI-faults & as well as the topic if we resolve this faults what could be happened to world.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    14 views3 pages

    AI Faults: Keywords:-AI AI-faults Risks of Ai Human Oracle

    Uploaded by

    International Journal of Innovative Science and Research Technology
    The paper is written to the development of the AI-faults, commonly known as AI-hard. Now the question arises should we resolve the faults? Is the faultsresolving beneficial for humankind? The paper suggests some important AI-faults & as well as the topic if we resolve this faults what could be happened to world.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    of 3

    Volume 3, Issue 3, March– 2018 International Journal of Innovative Science and Research Technology

    ISSN No:-2456-2165

    AI Faults
    Priya Dutta
    Computer science & engineering
    Tula’s Institute of engineering & management
    Dehradun, India

    Abstract:- The paper is written to the development of the II. EVALUATION OF AI


    AI-faults, commonly known as AI-hard. Now the
    question arises should we resolve the faults? Is the faults- The term AI used by john Macarthy, the father of
    resolving beneficial for humankind? The paper suggests AI, according to Wikipedia. In 1955 first AI program was
    some important AI-faults & as well as the topic if we designed. AI or Artificial Intelligence deals with the topic
    resolve this faults what could be happened to world. how one can evolve a machine which is able to think, learn,
    act, behave, and explain doubts like human can do. The
    Keywords:- AI AI-faults Risks of Ai Human oracle. machine should be then that kind of knowledgeable too. But
    the AI has faults, AI-faults suggest the lacking things or
    terms of AI. The evaluation of AI take place for mainly two
    I. INTRODUCTION reasons one is human’s need & another one is human’s
    greed. Need & Greed are quite similar sounds but they make
    Today’s modern era is based on machines & advance huge difference in practical circumstances.
    machines.
    As the entropy of a system is increasing we need to
    We want to ease our work, for this we require do huge amount of task. Tasks like fraud-detection,
    machines. Not only for easing a process for making it swift, automation, automating manufacturing/ support, safety
    adequate we are absurd of making advanced machines. But alerts, assembly line, introducing any kind of laws without
    the topic where we have to really determine is that this kind any corruption is easy for AI compared to humans. Humans’
    artificial-intelligence holding machines are really propitious. can get tired but a machine does not feel any tiredness.
    American writer Mr Elbert Hubbard writes, “One machine Machines can work any environment but human cannot, as
    can do the work of fifty ordinary men.” By doing fifty men’s in uncertain environment human live is more important than
    work alone a machine consume their chance of getting job a machine.
    as well as fabricating them idle. So in a simple perception an
    advanced machine is ruinous. Author Vernor Vinge righty In an interview with Larry King on his talk show,
    says, “Within thirty years, we will have the technological Scientist hawking said, “We are in danger of destroying
    means to create superman intelligence. Shortly after, the ourselves by our greed & stupidity.” Supporting his words
    human era will be ended.” His quote supports the real life humans countlessly purchasing AI; for not their need but for
    risk taking idea of making Ai fault free. A fault free AI their greed. He also visualized the competition which is
    supports understanding of natural languages, dealing with going to happen in near future with machines of humans.
    real life problems, knowing human emotions; in short
    producing machines which are capable of take the place of An AI includes machine learning, natural language
    human, easily. processing, expert system, vision, speech, planning, and
    robotics.
    AI- fault’s another term AI-complete was
    suggested by Fanya Montavo in the 1980’s, according to
    encyclopaedia.

    Fig. 1:- Artificial Intelligence include

    IJISRT18MA276 www.ijisrt.com 258


    Volume 3, Issue 3, March– 2018 International Journal of Innovative Science and Research Technology
    ISSN No:-2456-2165
    III. AI-HARD B. Natural language understanding
    Any kind of AI machine does not know any
    In computer language AI-hard symbolizes the complete natural language. The refor AI-fault free
    incompleteness of AI. To make an AI-fault free we need to system must understand natural languages.
    achieve the below goals: IV. REASONING
    A. Anti Captcha How to use Persevering knowledge at a need of
    As we often see accessing a webpage or making an time is called reasoning. An AI machine may have more
    online account we need to pass a Robot-test, where we have than knowledge to a common man but it does know how to
    to either select some images or digits, which distorted use this knowledge of its own.
    somehow, these instructions, are called captcha. This test
    could not pass by an AI machine until now because V. PROBLEM SOLVING
    machines cannot identify distorted characters.
    Problem search & solving capacity is still lacking in AI
    machines.

    These are the main AI-faults till now. Many


    companies like Google, Microsoft, Facebook, Apple, IBM,
    Deepmind, and Baidu are doing extreme effort to resolve
    this& making a complete AI.

    Fig. 2:- Source Google

    Problem Solving

    Things that make Natural


    Reasoning Language
    An AI complete
    Processing

    Anticaptcha

    Fig. 3:- AI-fault

    VI. AFTER FAULT RESOLVING SITUATION  Machine trucks took the place of normal trucks &
    drivers.
    After removing the faults AI system will became  Robots replace human in every work like journalism,
    more powerful & efficient. But what happens to the humans. accounts handling, computing, teaching and many
    more.
    The world has to face a human-machine face of. Where the  As we know there are so many partial men outside, thus
    creators are going defeat and created things are win. The
    making a sense may be there will be partial AI
    world has to face devastating economic problems, such as: machines also.

    IJISRT18MA276 www.ijisrt.com 259


    Volume 3, Issue 3, March– 2018 International Journal of Innovative Science and Research Technology
    ISSN No:-2456-2165
     Criminals can use to fulfil their motive producing a VIII. CONCLUSION
    negative AI machine which will do opposite to one’s
    good. Professor Moshe Vardi says, “We are approaching
     In machine learning, we cannot easily catch the culprits. a time when machines will be able to outperform humans at
    If a robot attempt any sort of wrong decision who will almost any task”. This kind evaluation is topic of debate
    take the blame? where some people oppose some supports the AI-hard free
     So many types ‘of privacy challenges will also be found concept. AI is developing fear &excitement both together.
    in near future because of AI. We are approaching an era where some uses the AI for
    doing & other doing harm for mankind. As the survey
    VII. SURVEY suggests people are not opposing the betterment but they
    also terrified about the upcoming strom.
    Name Address Occupation Opinion on
    Fault free REFERENCE
    AI
    [1]. Stuart J. Rusell Peter Norvig ‘s Artificial Intelligence,
    Mala Dhoolkot, House Wife If a machine
    A Modern Approach.
    Paswan Dehradun, can think &
    [2]. Elaine Rich, Kevin Knight, Shivashankar Nair ‘s
    Uttarakhand behave like
    Artificial Intelligence.
    us, then
    [3]. R.V Yampolskiy, Graphical CAPTCHA embedded in
    what the use
    cards, Western New York Image Processing Workshop
    of our
    (WNYIPW)-IEEE signal processing society, Rochester,
    brainsleft?
    NY, 2007.
    ProvakerRo Sitalkuchi, Student of Sometimes [4]. S.C Shapiro, Artificial Intelligence , in S.C.Shapiro, ed.,
    y coochbihar. B.A with it is helpful Encyclopedia of Artificial Intelligence, John Wiley,
    honours in to have an New York,1992,pp.54-57.
    English advance [5]. D.Shahaf and E. Horvitz, Generalized TASK
    sitalkuchi machine but MARKETS for Human and Machine Computation,
    college. it makes us Tweenty fourth AAAI Conference on Artificial
    more Intelligence, Atlanta, GA, July 2010.
    lethargic. [6]. M.A .Anusuya and S.K. Katti, Speech Recognition by
    Swagata Shanti para, Student of As a Machine: A Review, International Journal of Computer
    Sarkar Jalpaiguri,Wes IT in machine is Science & Information Security (IJCSIS),
    t Bengal. Siliguri stronger, 6(3)(2009),pp,181-205.
    Institute of durable than [7]. V.Bajaj, Spammers Pay others to Answer Security
    Technology humans’ & Tests, The New york Times, April 25,2010.
    . if it get [8]. Roma V, Yampolskiy, AI-Hard, orAI-Easy:
    human’s Classification of problems in Artificial Intelligence.
    brain it can
    destroy
    human’ as
    well the
    environment
    .
    Swagatam Korani para, Student of AI
    Dutta Jalpaiguri, class 12 in technology
    West Bengal FDI. somehow
    helpful but
    it can be
    harmful too.
    Debosnigdh Suhid lane , Student of Having an
    a Kundu Jalpaiguri, B.A with advance
    West Bengal. English machine
    honours in making our
    A.C work easy
    college. as well as
    efficient. It
    is not bad,
    having a
    robot-friend
    which
    thinks like
    us.

    IJISRT18MA276 www.ijisrt.com 260

    You might also like