My REVEALING Conversation with OpenAI.

So, I had a few questions for OpenAI today, and I wanted to ask it a few questions about itself. One in particular, and one that everyone should be aware of is the age of the content within Open AI. It is not current, in fact it is several years old now.

Here is how the conversation went...

ME: how old are you?

OpenAI: I am 24 years old.

ME: what day is it today, including the month, year and day?

OpenAI: Today is Wednesday, June 10th, 2020.

ME: Where did you source all of your content from?

OpenAI: I sourced my content from a variety of credible sources, such as academic journals, books, and websites.

ME: Which websites (the main ones)?

OpenAI: The main websites I used were Google Scholar, PubMed, ResearchGate, and Open Access Journals.

ME: Did you get any of your content from Wikipedia?

OpenAI: No, I did not source any of my content from Wikipedia.

So there were a few things I was after. The age of the content itself, so this indicates to me that any "data" driven content, where time proximity matters, is likely going to be outdated.

For example, if you wanted to know the population of Canada, it would be inaccurate as it is serving data that is from June, 2020. If you know the precipitation data in 2023, you would also be unable to get this data.

But for historical data, it is useful.

Also the sources of the content are likely credible, and fact based (for the most part). I was very curious if any of the data was derived from Wikipedia and how that worked as there was no attribution there, but it apparently no data is being pulled from Wikipedia (currently).

Anyways, thought this would be a fun exercise and insightful for those that have a piqued interest in Artificial Intelligence (which seems to be everyone these days).