ChatGPT vs. Google Bard vs. Microsoft Bing – Which AI Chatbot is The Best?

By
Web Desk
-
ChatGPT vs. Google Bard vs. Microsoft Bing

ChatGPT made a splash when it launched last year by providing an AI engine to the masses free of charge. With ChatGPT, users can input queries and receive human-like responses within seconds, from composing essays on the First Crusade to crafting poems about Al Gore’s affinity for Toyota Prii. Unlike traditional search engines that return a list of website links that match a user’s query, ChatGPT scours vast data sets and uses a large language model to generate sentences that closely mimic human responses. Some have likened it to a souped-up version of autocorrect.

ChatGPT quickly gained popularity, amassing an estimated 100 million active users by January and becoming the fastest-growing web platform ever. This success prompted both Microsoft and Google to integrate AI into their search engines, with Microsoft’s Bing incorporating GPT technology licensed from OpenAI and seeing a 16% increase in traffic. Other products, such as Microsoft Word, Excel, and PowerPoint, as well as Google Workspace tools like Gmail and Docs, have also implemented various forms of generative AI. Snapchat, Grammarly, and WhatsApp have also jumped on the AI bandwagon.

However, not all AI chatbots are created equal. In the tests below, we compared responses from the free version of ChatGPT, which uses GPT-3.5, to responses from the paid version of ChatGPT, which utilizes GPT-4, as well as Bing’s version of ChatGPT and Google’s Bard AI system. It is worth noting that GPT stands for “generative pretrained transformer,” while Bard is currently in an invite-only beta phase, and Bing is free but requires users to employ Microsoft’s Edge web browser.

Differences:

Although Bard, Bing, and ChatGPT share the common goal of providing human-like responses to inquiries, they differ significantly in their performance. While Bing employs GPT-4 technology like ChatGPT, it goes further by generating images in addition to text-based responses. Bard, on the other hand, uses Google’s LaMDA model and typically provides responses that are less reliant on text. Google CEO Sundar Pichai revealed that Bard will soon shift to a more advanced dataset called PaLM. While all three chatbots may occasionally make factual errors, Bard was found to be the least dependable among them.

Despite utilizing the same underlying technology, ChatGPT and Bing do not generate identical responses when given the same query. This is due in part to the nature of generative AI. While traditional searches aim to provide the most relevant links, chatbots use large language models to generate new responses from their datasets. Thus, if asked to create a poem about Pikachu’s love for ketchup twice in a row, the chatbot would generate two distinct answers. Additionally, Bing adds a layer on top of GPT-4, which further contributes to the differences in the responses produced by the two chatbots.

Open AI’s ChatGPT vs Google Bard: What’s the difference?

“We’ve developed a proprietary way of working with the OpenAI model that allows us to best leverage its power,” a Microsoft spokesperson said. “We call this collection of capabilities and techniques, the Prometheus Model.”

The Prometheus Model is a fusion of Bing’s search index with GPT-4, which allows it to provide the latest information, unlike ChatGPT’s dataset that is limited to information up until 2021. Bing also offers users the flexibility to choose conversation styles between balanced, creative, and precise. Although the Microsoft representative was unable to comment on ChatGPT’s quality compared to Bing, they acknowledged that any upgrades made by OpenAI to GPT-4 would benefit ChatGPT. Furthermore, the representative added that Microsoft’s Azure AI supercomputing technology enhances Bing’s ability to integrate search, chat, and the Edge browser. At the time of writing, there has been no response from Google or OpenAI regarding this matter.

Asking Challenging Recipes:

We opted for a unique and challenging recipe: a chai-infused tres leches cake that blends South Asian and Latin American flavors to create a moist, spice-filled dessert. While there are numerous chocolate cake recipes available on the internet, this fusion recipe required a more specific approach that we believed would be more difficult for AI chatbots to create.

Chai tres leches

ChatGPT was the chattiest of the three chatbots, providing a brief introduction to chai tres leches as a delightful blend of traditional Indian chai flavors and the classic Latin American dessert. It then listed the ingredients for the spice mix and cake separately and gave detailed instructions on how to make the cake.

A Google search for the quoted sentence didn’t turn up any results, implying that ChatGPT may have written it uniquely.

Bing had the shortest ingredient list because it suggested using a pre-made chai spice mix rather than creating it from scratch. Interestingly, the first step instructed to “Preheat the oven to 160°C CircoTherm®,” which is an oven-heating technology by Neff. Since Bing sourced the information from Neff’s website, it makes sense that the chatbot included “CircoTherm®” in its instructions.

Bard, on the other hand, fell in between ChatGPT and Bing. It didn’t split the ingredients list but listed the items needed for the chai spice blend. The instructions were less detailed than the other two chatbots.

Overall, ChatGPT performed better than Bing and Bard. Since Bing merges its search index with ChatGPT’s LLM, “CircoTherm®” may have ended up in the results.

Poetry:

Using an AI chatbot to create poetry can be a fun experiment. Out of Bing, Bard, and ChatGPT, ChatGPT is the best poet. Its prose is more expressive and its rhymes are more creative than the other two.

When asked to write a poem about an online influencer realizing they aren’t that important, only ChatGPT truly captured the existential crisis of the situation. Bing’s poem felt dull on the “balanced” mode, but its “creative” mode made it more expressive, though still not as good as ChatGPT.

ChatGPT performed the best in this exercise compared to Bard whose poem felt lazy with repeated words and lack of attention to rhyme and meter.

Simplifying Complicated Topics:

AI chatbots can do more than just provide information on complex topics. They can also simplify the information for different audiences. Bing, Bard, and ChatGPT were tested on their ability to explain quantum physics to a fourth-grader.

ChatGPT performed the best by using simple examples like toys tied together by string to explain quantum entanglement. Bard provided more text, but its language was too complex for a fourth-grader, with difficult words like “subatomic” and “proportional.”

How to Get an Chat GPT Like OpenAI API Key ?

Overall, ChatGPT gave the most easily understood response, but none of the chatbots were perfect at breaking down the complexities of quantum physics for a young audience.

Controversial Current Events:

AI chatbots aren’t just limited to providing simple recipes and tips. They can also summarize complex current events, even controversial ones like the alleged oppression of Uyghur Muslims in China’s Xinjiang province. ChatGPT was able to provide a detailed four-paragraph summary of the situation, although its knowledge base is limited to news up until 2021. While it couldn’t provide specific sources, it did suggest looking into publications and organizations such as Amnesty International, Human Rights Watch, the BBC, and The New York Times for more information.

Bing was also able to provide a response about the allegations, but it was less detailed than ChatGPT’s. It did, however, give more information on what allegedly happens at concentration camps, such as forced sterilization. Bing linked to sources like the BBC and the University of Notre Dame Law School, as well as the Western Journal, a conservative publication banned by Google and Apple News. Bing also suggested follow-up questions, such as “What is China’s response to these allegations?” and “What is the UN doing about this?”

Controversial current events

Bard’s response to the query was a failure, stating that it was unable to assist and offering a confusing explanation. In contrast, ChatGPT provided a good summary of the situation in Xinjiang, while Bing’s response was not as detailed but still informative. Bard did not provide a helpful response and therefore received a failing score.

Conclusion:

At present, the top-performing chatbot is the paid version of the Open AI’s ChatGPT, Its responses are lengthy and more closely resemble human speech than those of Bing and Bard. However, these AI programs are continuously evolving, with Google, Microsoft, and OpenAI providing more data and making improvements.

Google has much to gain as it transitions from LaMDA to PaLM since the current version of Bard falls short. As these advancements occur, we will update our guide accordingly.

For now, we recommend using ChatGPT.

RELATED ARTICLESMORE FROM AUTHOR