Chatbots aren’t new, but they certainly have new life. The first attempt at creating an interface allowing a computer to hold a conversation with a human dates back to 1966, when MIT professor Joseph Weizenbaum created Eliza. While that original chatbot matched user prompts to scripted responses, the development of natural language processing (NLP), machine learning (ML), and AI has helped these tools change how we communicate with software and, potentially, how we work, search, and acquire and distill information.

This guide explores the evolution of AI chatbots. It looks at the major players shaping the technology and discusses ways marketers can use the technology to engage audiences, customers, and prospects.

What is an AI chatbot?

Chatbots are AI systems that simulate conversations with humans, enabling customer engagement through text or even speech. These AI chatbots leverage NLP and ML algorithms to understand and process user queries. They can handle a wide range of tasks, from customer service inquiries and booking reservations to providing personalized recommendations and assisting with sales processes. They are used across websites, messaging apps, and social media channels and include breakout, standalone chatbots like OpenAI’s ChatGPT, Microsoft’s Copilot, Google’s Gemini, and more.

How do chatbots work?

Basic chatbots follow scripts and decision trees to provide canned responses. Most customer service-oriented chatbots used to fall into this category before the explosion of NLP. Those limitations gave chatbots a bad reputation, though. Salesforce’s 2023 Connected Financial Services Report found 39% of customers point to poorly functioning chatbots when asked about challenging customer experiences they encountered at their financial service institution.

On the other hand, AI-powered chatbots use NLP and ML to understand the context and nuances of human language as a knowledge base. They analyze user inputs to determine a user’s intent, generate responses, and answer questions that are meant to be more relevant and personalized. Over time, AI chatbots can learn from interactions, improving their ability to engage in more complex and natural conversations with users. This process involves a combination of linguistic rules, pattern recognition, and sometimes even sentiment analysis to better address users’ needs and provide helpful, accurate responses.

AI chatbots by the numbers

As chatbots’ applications in marketing, commerce, and more evolve, their impact on industry is wide and varied. Here are a few stats to consider:

  • Worldwide retail spending on chatbots is set to soar from $12 billion in 2023 to $72 billion by 2028, according to June 2023 data from Juniper Research.
  • By the end of 2024, 33.2% of adult consumers in the US will use AI-enabled banking chatbots, EMARKETER forecasts.
  • 49% of US adults have used an AI chatbot for customer service in the past 12 months, according to November 2023 data from Google and Ipsos.
  • Only 9% of adults in the US use chatbots daily to connect with businesses or service providers, according to Vonage Research.
  • 43% of CMOs and executives worldwide surveyed by Capgemini said they expect to frequently use chatbots for marketing in the next two to three years.
  • 50% of digital marketing agencies in North America reported using ChatGPT, according to a September 2023 survey by WordStream.
future channels used for marketing by executives worldwide to engage with customers
A chart showing future channels used for marketing by CMOs to engage with customers, October 2023. (Subscribers only)

What is ChatGPT?

ChatGPT is the chatbot that started the AI race with its public release on November 30, 2022, and by hitting the 1 million-user milestone five days later. Created by OpenAI, GPT stands for “generative pre-trained transformer.” It is designed to answer user questions, which can include simple queries for facts or figures or complex instructions for generating content and communications on behalf of the asker.

ChatGPT is part of a class of chatbots that employ generative AI, a type of AI that is capable of generating “original” content, such as text, images, music, and even code. Since these chatbots are trained on existing content from the internet or other data sources, the originality of their responses is a subject of debate. But the model essentially delivers responses that are fashioned in real time in response to queries.

Within a year, ChatGPT had more than 100 million active users a week, OpenAI CEO Sam Altman said at a developers conference in November 2023. EMARKETER forecasts ChatGPT will have 77.2 million US users in 2024.

ChatGPT’s user growth follows an equally rapid evolution of the platform since its debut. Its most recent release, GPT-4 Turbo, is already far more powerful than the GPT-3.5 model it launched with. It has since rolled out a paid tier, team accounts, custom instructions, and its GPT Store, which lets users create their own chatbots based on ChatGPT technology.

Generative AI chatbots: The competition

ChatGPT may have started the AI race, but its competitors are in it to win, which isn’t surprising since many of them are the most influential tech companies in the world. Here are a few chatbots competing for users.

Google Gemini (formerly Google Bard)

Google isn’t used to being behind, but since the launch of ChatGPT it has been racing to catch up to, and possibly overtake, the OpenAI leader.

Its first chatbot, Bard, was released on March 21, 2023, but the company released an upgraded version on February 8, 2024, and renamed the chatbot Gemini.

Gemini uses AI models developed by Google. Like ChatGPT, it can handle a wide range of multimodal queries, which means it can process text, generate images, and work with audio files. In addition to getting its own Android app, Gemini will also be integrated into other Google applications like Gmail and YouTube.

Claude 3

Claude 3 is a generative AI chatbot created by Anthropic, a company founded by several former OpenAI employees. This third iteration of the chatbot was made available to the public in March 2024.

Claude has similar functionality to ChatGPT and Gemini. It can respond to text-based queries and generate a range of content on-demand. However, Claude is different in that it goes beyond its competitors to combat bias or unethical responses, a problem many large language models face. In addition to using human reviewers, Claude uses “Constitutional AI,” a model trained to make judgments about outputs based on a set of defined principles.

Chatbot web search experiences

The rise of AI chatbots is also primed to remake the way consumers search for information online. In addition to offering a more natural and conversational interface, AI chatbots can summarize information and create original, generated responses that are based on internet sources but don’t deliver the usual list of links that today’s search engine results pages offer. Here are a few options that are emerging.

Google’s Search Generative Experience (SGE)

Google might be disrupting itself with SGE, but the company is making sure that it maintains its search dominance if chat interfaces become the norm. Google’s SGE creates generated results based on search queries, linking to sources it uses in the response in the form of contextual links or thumbnails.

Microsoft Copilot (formerly Bing Chat)

Microsoft’s Bing search engine is also piloting a chat-based search experience using the same underlying technology as ChatGPT. (Microsoft is a key investor in OpenAI.) Microsoft initially launched its chatbot as Bing Chat before renaming it Copilot in November 2023 and integrating it across Microsoft’s software suite.

Though Microsoft was first to release a chatbot search experience, it has not made a big dent in Google’s market share, which holds at 91.6% compared with Bing’s 3.3% market share, according to February 2024 data from StatCounter.

Perplexity

Perplexity AI is a chatbot that is aimed at replacing traditional search. While not so different from other chatbots, this “answer engine,” as the founders describe it, generates answers to queries by searching the internet and presenting responses in concise, natural language. Unlike Google and Microsoft, which are experimenting with integrating ads into their search experience, Perplexity aims to stay ad-free.

Arc Search

Launched in early 2024, Arc Search is a standalone mobile search app created by The Browser Company, which also owns the Arc browser. Its app can “browse” for users based on queries and generates unique results pages that act like original articles about the topic, linking to all of the sources it uses to generate the result. Like Perplexity, the service does not include ads, and the Arc browser connected to it even blocks web trackers and on-page ads by default.

How marketers can use chatbots

Chatbot marketing is the broad term for encompassing the use of chatbots across the web, apps, social platforms, messaging, and more to converse with customers in order to sell products and services or to inform customers about key service information.

Many marketing chatbots are deployed on platforms such as Facebook Messenger, WhatsApp, WeChat, Slack, or text messages. However, the rise of conversational AI has expanded the range of chatbot tools, as well as how naturally they interact with customers.

how US B2B marketers are utilizing AI
A chart showing how US B2B Marketers are utilizing AI.

Engaging customers through chatbots can also generate important data since every interaction improves marketers’ ability to understand a user’s intent. The more successful chatbots are the ones that are able to drive a good conversational experience with human-like responses.

Here are a few ways marketers and retailers can employ chatbots:

  • Use chatbots instead of online forms to capture information for lead generation campaigns. In fact, follow-up questions posed by the chatbots can help with lead qualification too by giving marketers more information about a customer.
  • Generating text and other content creation to engage customers about their loyalty programs, reminding users about their points, tallies, and more.
  • Offer customers easy ways to get quotes about products and services and seamlessly transition to the order or reservation process.

Chatbots are also starting to be integrated into the ad creation process. Google, for example, has released a chatbot powered by Gemini that helps advertisers create ad copy and creative through a chat-based interface.

Limitations and risks of chatbot marketing

As we pointed out at the beginning of this guide, customer experience with chatbots hasn’t been serendipitous for most people. Clunky, intrusive experiences and frustrating interactions have marred the medium, but integration of AI in chatbots aims to smooth out a lot of the wrinkles companies have had with building affinity for chatbots.

AI, however, brings its own risk. While ensuring that responses are free of bias and brand safety are essential, chatbots still struggle with delivering accurate information and are prone to “hallucinate,” making up answers that are patently false.

For example, Google’s Gemini and Microsoft’s Copilot both were found to give out inaccurate information about Super Bowl LVIII in February 2024. The mistakes ranged from naming a winner before the game even happened to misstating player stats.

How to build your own chatbot

For marketers looking to engage in chatbot marketing, there are a host of avenues. Native messaging apps like Facebook Messenger, WeChat, Slack, and Skype allow marketers to quickly set up messaging on those platforms. Of course, generative AI tools like ChatGPT allow marketers to create custom GPTs either natively on the platform or through API access. Meta offers a similar option through its AI Studio.

There are also a number of third-party providers that help brands get chatbots up and running. Some of those services are free, such as HubSpot’s chatbot builder, while companies like Drift and Sprinklr offer paid chatbot tools as part of their software suites.