Change % (1H)
Change % (24H)
POLY AI (AI) Historical Price & Volume Charts
POLY AI Wikibeta
|Mkt. Cap||USD 114||Volume 24H||USD 0|
|Mkt. Share||0.00 %||Available Supply||2,281,981|
|Change % (1H)||-0.50 %||Max Supply||0|
|Change % (24H)||-32.53 %||Total Supply||29,000,000|
|Change % (7D)||14.66 %||Proof|
|Algorithm||Upated: 6 months ago|
POLY AI Historical Data
POLY AI Videos
Artificial Intelligence Coin Grading - AI Coin Grading For The Future
Effect Ai ICO Review - King of the AI Coins? (vs. SingularityNET, DeepBrainChain, and Gems)
POLY AI (AI) Reviews & Critics
We need a more consistent grading system with tolerances and allowances that help define each grade. Every profession like engineering, fabricating and even ....
- Also say someone did buy a fraudulent coin like from that guy on etsy selling a penny for $50,000 the ai baseline would help them represent their case say that person ever had to go to court. 10-50 grand lost on a fake coin isn't a small matter even for people who are a bit weathlier.
- Very insightful. I think it would help eliminate fraud too. Especially would be helpful to someone new to coins like me. Wouldn't have to be as concerned about getting a fare score cause there'd be that instant baseline set right off the bat.
- The theory behind having a computer determining a grade is a great one. How it gets the baseline knowledge is the problem with this system. Let's use your example of the 2013 Kennedy half. In your explanation you stated that your coin was graded a 67 and that you have seen others both better and worse with the same grade. A computer is never better than the information it is given so if you were to give the computer a sample size of 7000 coins, specifically this 2013 kennedy half in a 67 grade, a percentage of them will not be 67's due to the explanations that you gave in the video. So, you are using a baseline that is faulty so some coins "could" be culled that were actually 67's and other might be a weak 67 grade or maybe even a 66 because the data would be faulty. The one difference here is that true AI has the ability to learn from it's mistakes. There would have to be thousands of "test runs" with the AI so that it's knowledge base could grow. This is only one coin grade. Then you would have to establish a base of information from other graded of lesser value numerically. This means that if you are using 7000 67's, to keep the system even you must include 7000 66's. My final point to make here you could have had 30 different graders since 2013. Let's just say 1/3 of them have their own "niche" in their grading style we would again be sending data to the computer of which a percentage would be flawed. Now we are just talking in this instance many thousands of coins for Just the 67's and the 66;s. On the other side of this what do we do with coins that have only few examples, If my memory serves me correctly there is only one ms69 morgan dollar. At lease 3 judges make that original determination. We know that, in general, the higher grade, especially in older coins, not many of each example exists. As I said at the start, the theory is, generally, sound. The practicality of doing this I would say is "not so much." because you are already dealing with factors such a multiple judges for the same coin and if the coin was a 'break out' coin, then you maybe dealing with as many as 6 or more people's opinion before the information ever gets to the computer to be uploaded to the AI.
- I agree!