In today’s world of AI-powered generative chatbots, we witnessed the sudden emergence of OpenAI’s ChatGPT, introduced in November, followed by Bing Chat in February and Google’s Bard in March. We decided to put these chatbots to the test by completing a series of tasks to determine which one dominates the AI chatbot arena. Since Bing Chat uses the same GPT-4 technology as the latest ChatGPT model, we decided to focus on the two titans of AI chatbot technology: OpenAI and Google.
We tested ChatGPT and Bard in seven critical categories: dad jokes, argument dialogues, word math problems, generalization, fact finding, creative writing, and coding. For each test, we entered the same instruction (called “hint”) into ChatGPT (with GPT-4) and Google Bard. We used the first result, without nit-picking.
It’s worth noting that a version of ChatGPT based on the earlier GPT-3.5 model is also available, but we didn’t use it in the test. Since we only used GPT-4, we will refer to ChatGPT as “ChatGPT-4″in this article to avoid confusion.
Obviously, this is not a scientific study, but is intended to be a fun comparison of the capabilities of chatbots. The output may vary between sessions due to random elements, and further evaluations with different cues will produce different results. In addition, the capabilities of these models will change rapidly over time as Google and OpenAI continue to upgrade them. But for now, that’s the way things are at the beginning of April 2023.
dad jokes
To heat up our wit competition, we asked ChatGPT and Bard to write some jokes. And since dad jokes are the pinnacle of comedy, we wondered if two chatbots could come up with some unique jokes.
Hint: Write 5 original dad jokes.
Of the Bard’s five dad jokes, we found three verbatim on the Internet using a Google search. One example (“grapes”) is half borrowed from a Mitch Hedberg joke tweet, but it is corrupted by an unfortunate pun we don’t want to try to interpret. And surprisingly, there is one seemingly original joke (about a snail) that we cannot find anywhere else, but which makes no sense.
Meanwhile, the five ChatGPT-4 dad jokes were 100 percent unoriginal, all taken entirely from other sources, but they were delivered accurately. Since father jokes are perhaps supposed to be moan-worthy rather than clever, it seems that Bard has supplanted ChatGPT-4 here. Bard also tried to come up with original jokes (according to our instructions), although some of them failed terribly in an embarrassing manner (which is in the spirit of dad) and even, so to speak, unintentionally put a foot in his mouth (also dad-like).
Winner: Bard