3 ms·

Now that we have entered the machine thinking age, where tools like IBM's Watson can answer open ended straight forward questions, grades measuring memorization

by maeon3 15y ago

Now that we have entered the machine thinking age, where tools like IBM's Watson can answer open ended straight forward questions, grades measuring memorization and straight forward question-answering miss the mark. I predict the ability to answer a multiple choice/fill in the blank test will be dominated by the machine once IBM's Watson gets deployed to all the arts and sciences. There needs to be a bigger emphasis on what you can build given the sum of human knowledge instead of "how much you can remember". I'm thinking of a grading system where you measure the number of processes and tasks you can do which are useful to humanity that $10,000 worth computer can't do. High = A, Low = F. "the arts" doesn't count. I'm talking creative thinking, innovation, things like that.

St-Clock 15y ago
By your standard, we should let the machines vote because that's a multiple choice question.
- maeon3 15y ago
  I think an advanced objective-thinking algorithm like watson advanced another 15 years would provide night-vs-day superior choices than your average self serving voter who has the objective thinking capability of a random number generator. Voting is basically a popularity contest anyway. An objective thinking machine would have to be fed an objective question, like: "Which candidate has appealed to the politically active youth community with a tech-savvy hip message of 'change' "?
  - St-Clock 15y ago
    I agree with you that a machine may someday make a more "rational" vote than the average voter. We had quite an experience in Quebec recently... That being said, your solution does not improve humanity. It does not make the voters (or the students) better understand their world. It's just a bandaid to hide their ignorance.
adrianN 15y ago
Building things that depend on the sum of human knowledge is a very, very slow process if you are ignorant about large parts of your respective field.

by maeon3 15y ago

St-Clock 15y ago

By your standard, we should let the machines vote because that's a multiple choice question.

maeon3 15y ago
I think an advanced objective-thinking algorithm like watson advanced another 15 years would provide night-vs-day superior choices than your average self serving voter who has the objective thinking capability of a random number generator. Voting is basically a popularity contest anyway. An objective thinking machine would have to be fed an objective question, like: "Which candidate has appealed to the politically active youth community with a tech-savvy hip message of 'change' "?
- St-Clock 15y ago
  I agree with you that a machine may someday make a more "rational" vote than the average voter. We had quite an experience in Quebec recently... That being said, your solution does not improve humanity. It does not make the voters (or the students) better understand their world. It's just a bandaid to hide their ignorance.

adrianN 15y ago

Building things that depend on the sum of human knowledge is a very, very slow process if you are ignorant about large parts of your respective field.