Bard vs. ChatGPT: What’s the difference? [2024]

Bard vs. ChatGPT: What’s the difference? [2024]

We’ve all witnessed the hype around ChatGPT since OpenAI released it to the public. Chances are you’re one of its 100 million users—I know I am. I’ve used the AI chatbot to generate content ideas for articles, write emails I’d rather not write, and get Google Sheets formulas I can’t figure out on my own. (Which is often.) 

While I’ve perfected the art of writing prompts to get the best result, I’m painfully aware of ChatGPT’s limitations. For example, only paid subscribers can access the more powerful GPT-4, which can browse the web and generate AI images. 

Bard, on the other hand, offers a lot of what ChatGPT does, faster and for free. But there are plenty of other features that set them apart. Here are the main differences I discovered while comparing Bard vs. ChatGPT. 

ChatGPT vs. Bard at a glance

At a base level, both chatbots use natural language processing, which means users key in a prompt or query, and the chatbots generate a human-like response. 

There are a few key differences, though, that boil down to the models they’ve been trained on and access to the AI chatbot itself (Bard is available for free in more than 40 languages and over 230 countries and territories; ChatGPT offers nine languages and is available in 164 countries and territories). 

I’ll walk through some of the core differences between ChatGPT and Bard in depth in the coming sections, but here’s a quick breakdown of how they compare.






Language model

OpenAI’s GPT-3.5 (ChatGPT Plus: GPT-4)

Pathways Language Model (PaLM 2)

Data sources

ChatGPT was trained on a massive dataset of text, including Common Crawl, Wikipedia, books, articles, documents, and content scraped from the open internet; sources for the free GPT-3.5 model end in 2021, but ChatGPT Plus users can use GPT-4 to perform web searches

Bard was trained on Infiniset, a data set including Common Crawl, Wikipedia, documents, and conversations and dialogues from the web; Bard can also perform web searches


Free; ChatGPT Plus is available for $20/month


Bard offers a better user experience overall 

At first glance, Bard and ChatGPT offer similar features: 

  • Text formatting. Both AI chatbots automatically apply basic formatting—like headings and bolded text—to their responses, making it easier to scan through. 

  • Chat history. All conversations are automatically stored and managed in the side panel of each chatbot’s home page for easy reference. Both chatbots also allow you to turn off saving your chat history entirely. 

  • Quick-action buttons. You can like or dislike responses (to inform future responses) and share conversations with the click of a button. 

  • Data sources. Bard can provide a list of relevant sources to its text responses if prompted. Interestingly, though, Bard will automatically link an image pulled from the web to its original source. ChatGPT can only cite sources if it’s running on the web-browsing version of GPT-4. If you ask ChatGPT using the GPT-3.5 model to cite its sources, it tries to be helpful by providing you with a list of resources to verify its response. For example, when I asked it to cite its sources about kestrels, it suggested I refer to “reputable field guides on birds of prey, ornithology textbooks, or websites dedicated to birdwatching and bird identification” to learn more. 

Beyond that, here’s where Bard and ChatGPT differ in terms of user experience. 

Bard makes it easy to modify questions and responses 

If you want to modify a response in ChatGPT, the only way to do it is to prompt ChatGPT again with specifics on how you want it to update its answer. 

Or you can add custom instructions to ChatGPT, so it automatically considers certain preferences (like a shorter response versus a longer one) without you typing it out in the prompt every single time. 

Bard, on the other hand, lets you quickly modify every response by length or tone without needing to enter a new prompt. All you have to do is click the Modify response icon, which looks like a stack of horizontal bars, and select one of the available modifications: shorter, longer, simpler, more casual, or more professional. 

Quick-action buttons in Bard that let you quickly modify Bard's response by length or tone.

ChatGPT and Bard can browse the web, but Bard lets you do it for free 

ChatGPT and Bard can perform web searches to inform their responses, but there are a couple of differences. 

  • Bard lets all users access the internet, whereas ChatGPT limits this function to users on a paid subscription. 

  • When I used the chatbots to browse the web, ChatGPT took significantly longer than Bard to generate a response, and ChatGPT regularly ran into network errors.  

ChatGPT offers back-and-forth voice dialogue on the mobile app 

OpenAI recently launched the ability for ChatGPT Plus and Enterprise users to verbally communicate back and forth with ChatGPT. You can even change ChatGPT’s voice—the voice options were created using voice actors, so they all sound really natural. 

A list of ChatGPT's voice options in the mobile app.

ChatGPT’s voice capability is a really useful feature that makes it easier to interact with ChatGPT—but it’s currently available only on iOS and Android. And it doesn’t work in tandem with ChatGPT’s web browsing feature. This means that if you use the voice function, ChatGPT’s responses will be limited to data prior to September 2021. 

Bard doesn’t offer back-and-forth voice conversations (yet). And while it can speak its responses out loud, its robotic voice leaves much to be desired. 

Example of a Bard response with the option for Bard to voice its response highlighted.

Bard is built for research, whereas ChatGPT is the better writer 

Bard’s one very chatty assistant, zealous in its approach to gathering research for you to make your life easier. Does it cite its sources as it goes? No. But it will if you tell it to. 

Or you can click Double-check response (the Google icon) under any response to learn more or fact-check Bard’s response. (Text highlighted in green indicates links to Google Search results with similar information; text highlighted in orange indicates results with differing information.) Bard will even offer related search queries on a given topic, which can be helpful when you’re doing any level of research. 

Example of a Bard response with select parts of its response highlighted, along with a list of related search topics.

GPT, on the other hand, is the industry standard when it comes to natural language tasks, powering other AI tools like Jasper,, and Bing’s AI tools. It’s highly trained on web text and more focused on generating text based on statistical patterns. As part of ChatGPT, it functions as a chatbot, but it can also serve as a summarizer, a translator, and other roles on a more textual level. 

This makes GPT (in my opinion) a better writing tool than Bard. As a quick example, when I asked ChatGPT to write a tweet about Zapier, it did so pretty much in line with the rules for writing tweets (e.g., short character count, emojis, and hashtags).

A post about Zapier written by ChatGPT.

Sure, it’s a little generic, but I could share this tweet (or post) as is, and it’d work.  

Bard, on the other hand, gave me only the elements of a post: a message, hashtags, and emoji. It’s almost as if it wanted me to choose my own adventure. 

A post about Zapier written by Bard.

In both cases, a little prompt engineering would drastically improve the output, but this gives you a sense of how the two chatbots operate differently. 

I also found ChatGPT to be better at brainstorming blog ideas, writing long-form articles or business emails, and coming up with content marketing ideas. For example, I asked both AI chatbots to create an outline for a blog post about the crisis surrounding bees. 

Here’s a portion of the outline from Bard. 

Outline for a blog post about the crisis surrounding bees written by Bard.

And here’s a portion of the outline from ChatGPT. 

Outline for a blog post about the crisis surrounding bees written by ChatGPT.

While Bard and ChatGPT both generated solid outlines, ChatGPT took it one step further by also drafting a working title, headings, and subheadings—all of which I find to be the trickiest part of writing articles. 

Bottom line: while GPT can understand and generate a wide range of text for multiple purposes, including content marketing, Bard feels like it was designed primarily to act as a research tool. 

ChatGPT and Bard make it easy to share conversations, but Bard takes it one step further by connecting to Google apps

Both ChatGPT and Bard allow you to share your conversation with others. Anyone who has access to the conversation can even pick up where you left off. 

But here’s where ChatGPT and Bard differ. 

ChatGPT doesn’t let you share conversations with images 

If you upload an image to a chat with Bard and then share the entire chat, the image will be visible to and downloadable by anyone who has access to that conversation. 

If you upload an image to a conversation with ChatGPT, you’ll have to keep it to yourself. Or you can copy or screenshot the response. Sharing conversations with images is not yet supported. 

Bard lets you export responses to Google Docs and Gmail 

Since Bard is a Google product, it’s no surprise that it’s connected to your Google Workspace—specifically, Google Docs and Gmail.

Let’s say you used Bard to create an article outline. Now you can export that response to Google Docs and begin drafting—no copying and pasting required (unlike ChatGPT). You can do the same thing with Gmail, too. 

How to export a Bard response to Google Docs or as a draft message in Gmail.

Bard can retrieve images from the web; ChatGPT can generate AI images

Bard can surface relevant images from Google Search, which is a key feature that sets it apart from ChatGPT. Whether you’re researching specific dog breeds, images from the James Webb telescope, or even bicycle repairs, Bard can churn out specific images from other web pages for visual context. 

A response from Bard that includes relevant images pulled from the web.

You can also click on an image, and Bard will open the web page with the image source in a new browser tab. 

While ChatGPT can’t retrieve images from the web, it can generate AI images using DALL·E 3. (Of course, this feature is limited to users with a paid subscription.) This means you can use ChatGPT to do things like generate blog images and create business logos. Or you can use it for what it was clearly intended to do: create museum-worthy images of a fluffy dog painting a vast array of galaxies.  

ChatGPT conversation with four AI-generated images of a Pekingese dog wearing a beret, sitting on a stool beside a painting of galaxies.

Bard and ChatGPT can analyze images with prompts, but Bard lets you do it for free 

You can also upload images with your prompt and ask Bard to analyze them for you. For example, you can ask Bard to caption an image. 

A list of image captions generated by Bard.

ChatGPT Plus and Enterprise users can do the same thing with ChatGPT, but again, you can’t then share that conversation—or any ChatGPT conversation with images—with other people. 

An image caption written by ChatGPT.

ChatGPT has more powerful integrations

Bard and ChatGPT can both connect directly to other apps, but they go about it a bit differently.

Bard uses Bard Extensions, which retrieve real-time information from other Google apps, including Gmail, Google Drive, Hotels, Flights, Maps, and YouTube. This means you can do things like ask Bard to find cheap flights for your next trip or pull details from an email confirmation in Gmail—all without leaving the chat. 

Here’s an example of Bard retrieving flight options for an upcoming trip. 

Bard retrieving flight details for a flight from Toronto to Cancun.

ChatGPT does something similar, but it uses plugins. There are over 30 plugins to choose from on ChatGPT, and you can get them to work in tandem. For example, you can ask the AI for a recipe recommendation, get an accurate count of calories (using the Wolfram plugin), and then ask it to create a shopping list (with the Instacart plugin). 

Example of a ChatGPT response generated in tandem with two ChatGPT plugins.

And this isn’t the only way that ChatGPT extends beyond the Google ecosystem. With Zapier’s ChatGPT integration, you can connect ChatGPT to thousands of other apps and build ChatGPT directly into your existing workflows, regardless of your tech stack. Learn more about how to automate ChatGPT, or take a look at these workflows to get started. 

To get started with a Zap template—what we call our pre-made workflows—just click on the button. It only takes a few minutes to set up. You can read more about setting up Zaps here.

Bard vs. ChatGPT: which is better?

That was a lot to digest, so let’s do a quick rundown of each AI chatbot’s pros and cons.

Google Bard: pros and cons



Bard offers a better user experience overall, complete with the ability to easily modify prompts, export responses to Google Docs and Gmail, and share conversations

Bard can read its responses out loud, but its robotic voice leaves much to be desired 

Bard can perform web searches faster than ChatGPT and offers it for free to all users

While Bard provides sources and external links to resources (if prompted), the sources aren’t always reliable 

Bard can retrieve images from the web

Bard provides a fairly isolated experience, with no plugins or integrations other than with Google apps

ChatGPT: pros and cons



ChatGPT is better at generating text, such as long-form articles and emails, and can generate AI images 

Web access is available only to ChatGPT Plus and Enterprise users 

ChatGPT offers a collaborative experience, with the ability to share conversations with others 

Web searches can be glitchy and run much slower than Bard 

ChatGPT has a whole suite of plugins (and a Zapier integration) that offer more use cases with different apps

ChatGPT doesn’t let you share conversations with images 

It’s worth mentioning that ChatGPT and Bard share a noteable con: both chatbots are prone to generating plausible-sounding but inaccurate responses (also known as hallucinations).

At the end of the day, the better AI tool depends on what you’re using it for—and whether you can deal with those pesky hallucinations. 

Related reading:

This article was originally published in March 2023 by Elena Alston. The most recent update was in October 2023.

Leave a Reply

Your email address will not be published. Required fields are marked *