I wanted to see the impact of  AI on  KnifeDB.com.  After all, if you can just tell AI to find a price for any specific Case pocket knife, then there is no sense in my continuing to add knives and prices while attempting to make a powerful price guide for collectors.  I first asked how much a 1970 5254 is worth.  The answer was not bad: it talked about condition and gave a  general range of prices.  As I worked two different AI programs, this became clear to me.  The AI programs can provide general answers but are not good at specifics.  For example, I told ai to show a picture of every 2016 muskrat with prices.  It showed 4, while I have almost 40 documented.  And the prices were general, and not attached to any of the four knives it put forward.  I asked for all 2015 sets and it showed some of them, but  not all.  So I have decided to continue my work on Case xx knife values.  The following table may be of interest to collectors.  It shows the number of knife variations currently documented by years.  I am currently working on 2016, and you can tell that the years after that still need most work done.  The very early knives (before 1915) also need nore focus.  But between 1915 and 2016 I would put KnifeDB at a 90+ percent data completion level.  There are over 30,000 Case XX pocket knives in the database.  Most have at least one actual selling price.  And these are adjusted twice a year for inflation to give the serious Case XX the most accurate price information avialable anywhere.  Period