Over a million people have now accessed the ChatGPT AI (by OpenAI), an advanced AI chatbot that extends far beyond dictionary answers. It uses a vast library of text information (data) that is used to train it, allowing it to recognize patterns and generate answers based on this. Moving beyond “if this, then that” statements, the AI is capable of returning unique answers each time a question is asked and can answer some pretty complicated requests.
One could ask it to write a simple plugin for their website or provide a working example of code for their application. One could also ask it to write their high school essay project for them (not that you would do that, right?) or to cook up a much better description of something than what their mind was capable of (maybe you didn’t get enough sleep last night).
It’s answers can sometimes move far beyond a typical response you’d get out of something like Amazon Alexa or Google Assistant. Take the following random math question I asked it. See how it didn’t just say “the answer is….”. Instead, it provides a full conversation on the topic.
You literally are having a conversation with the AI and not just a cold question->answer flow that you’d find with the previously mentioned digital assistants.
I have spent a lot of time playing with it and have had some wonderful results. I even asked it to write me a WordPress plugin that I always wanted for a personal blog (but never got around to it) and it actually did (mostly). I had to finish it off by adding a “);” while finishing off the end of the PHP script (a final “}” and “?>”) and writing the second half of the script (it created the part that manages it on the backend but not the part that displays it on the front end for whatever reason). So it actually wrote usable PHP script that didn’t take but a few seconds to sum up to a working file (for it’s half).
Then I asked it to write me a short story. But not just any short story. I wanted to provide it with an odd request that may be a little hard to wrap it’s digital mind around. However, I apparently didn’t try hard enough as it immediatly started typing the response and had a short story written in seconds. It even gave my main character a name.
The request: “Write a short story about a transdimensional bipolar tyrannosaurus rex that likes to wear fruit hats, speaks over 11 languages, dreams about being a professional boxer but can never make it due to his short arms, and dominates at Dance Dance Remix”
It’s response:
Once upon a time, in a far-off land on the other side of the multiverse, there lived a transdimensional bipolar tyrannosaurus rex named Timmy. Timmy was quite a unique creature, with a love for fruit hats and the ability to speak over 11 different languages. He was also a dreamer, often fantasizing about one day becoming a professional boxer.
However, Timmy was not without his challenges. Despite his impressive size and strength, he was plagued by his short arms, which made it difficult for him to throw a punch with any real force. Despite his best efforts, Timmy was unable to fulfill his dream of becoming a boxer.
But Timmy was not one to let his setbacks get him down. He quickly discovered a new passion: Dance Dance Remix. With his incredible rhythm and quick feet, Timmy quickly became a force to be reckoned with on the dance floor. He would often spend hours practicing his moves, perfecting his craft and dominating the competition.
Despite his love for dancing, Timmy never forgot his dream of becoming a boxer. He continued to train, determined to overcome his short arms and achieve his goal. And while he may never have become a professional boxer, Timmy’s love for dance and his never-give-up attitude made him a true champion in the hearts of all who knew him.
That response was absolutely fantastic, catching me completely off guard (by a bit). Sure, it just wrote a bunch of fully functioning PHP and answered a bunch of a complex math questions I threw at it. However, this story required a certain level of creative imagining and it didn’t skip a beat when spitting this answer out.
Now looking back at what I was discussing a moment ago. Imagine if this kind of AI was available to us in our daily lives. As accessible as something like Amazon Alexa or Google Assistant. You can be sure that it will happen eventually. The question is, will it be this useful or will it be scary?
The biggest limitation affecting the ChatGPT AI is that it doesn’t have direct access to the internet. You have access to talking to it over the internet but it can’t go out and surf anything outside of its own database of information that is used to train it. It appears to be quite the massive database, but it is no internet. The moment you allow an AI like this access to the internet, the tables can change quickly. As multiple tech giants have discovered multiple times over the past five or so years, leading to the plug being pulled on these AI creations each time. Like Microsoft’s bot that it released on Twitter and quickly became racist in less than a day as it learned from all the users it interacted with.
Regardless, this technology will eventually be expanded upon and eventually injected into everyday tech. So many companies have been investing in AI development which has hit the fast-forward button on the evolution of it and related technologies.
OpenAI (itself) is a research company that was originally founded by Elon Musk, Sam Altman, Ilya Sutskever, Greg Brockman, Wojciech Zaremba, and John Schulman. It’s sole purpose is to develop friend AI solutions that could benefit everyone. It has since received major donations by other companies, such as Microsoft. It has been making major leaps in AI tech, including the DALL-E image generator (and, of course, ChatGPT).
What’s nice, is you don’t have a bot that is turning racist or authoritarian this round. Showing that OpenAI really is working on magic. This isn’t the first chat bot that has come from its research and it won’t be the last. Of things continue to get better, it may very well have the answer to releasing a friendly solution for the general public that can indeed be included in our daily-use digital assistants (or at that point, home AI devices). It won’t be long until we have our own personal Jarvis at home.
What do you think about the incredible accuracy and creativity that ChatGPT has been showing the world? Do you think this is going to pave a path toward much great technological evolution? Or do you think this is going to end badly (ie, Skynet) for everyone? Feel free to share your own thoughts in the comments below.
