Chances are, whether you’re scrolling through online stores, seeking information, or simply exploring the web, you’ve encountered a chatbot. But traditional chatbots from the BC (Before ChatGPT) era lacked the smarts to truly engage users.
In most cases, closing a chatbot became just an annoying step (right after accepting cookies) when you land on a website.
But now ChatGPT is Making Chatbots Great Again!
But how did we get here?
Let’s go on a journey through the history of chatbots, from their humble origins to the cutting-edge capabilities of ChatGPT.
1960s - ELIZA: The very first chatbot
Our story begins in 1966 with the creation of ELIZA (short for “Electronic Linguistic Analyzer”), perhaps the world’s first chatbot. Developed by Joseph Weizenbaum at MIT, ELIZA was a rudimentary program designed to mimic a Rogerian psychotherapist.
Weizenbaum’s goal was to explore the possibilities of natural language processing and human-computer interaction by creating a program that could engage users in conversation.
One of ELIZA’s most famous implementations was the “DOCTOR” script, which simulated a therapy session between the user and a virtual psychiatrist. Users would type in statements or questions, and ELIZA would respond with pre-programmed replies that reflected back the user’s words in the form of questions or prompts for further elaboration. For example, if a user typed “I feel anxious,” ELIZA might respond with “Why do you feel anxious?”
The result was a surprisingly engaging conversational experience that led many users to believe they were interacting with a real human therapist. People were drawn to the idea of confiding in a non-judgmental digital confidant, even if they knew on some level that ELIZA was just a cleverly programmed computer program. Here’s an example of a typical interaction with ELIZA:
User: “I’m feeling anxious about my upcoming presentation.”
ELIZA: “Tell me more about your feelings of anxiety.”
User: “I’m worried that I’ll forget my lines and embarrass myself.”
ELIZA: “Why do you think you might forget your lines?”
ELIZA’s impact extended far beyond its initial incarnation. It inspired generations of researchers and developers to explore the possibilities of conversational AI, laying the groundwork for future advancements in natural language processing and human-computer interaction.
1970s - PARRY: Simulating paranoid schizophrenia
In the early 1970s, amidst the burgeoning field of computer science and artificial intelligence, another pioneering chatbot emerged: PARRY. Developed by psychiatrist Kenneth Colby at Stanford University, PARRY was designed to simulate a person with paranoid schizophrenia, providing researchers with a tool to explore the intricacies of human psychology and behavior.
PARRY’s creation was inspired by the work of ELIZA. However, while ELIZA focused on engaging users in open-ended conversations, PARRY had a specific personality and backstory—a patient suffering from delusions of persecution.
What set PARRY apart was its ability to engage in remarkably lifelike interactions, drawing on a vast database of pre-programmed responses and scenarios. Users could converse with PARRY about a wide range of topics, from everyday small talk to more serious discussions about its perceived enemies and conspiracies.
One of PARRY’s most notable achievements was its ability to convincingly emulate the symptoms of paranoid schizophrenia, including delusions of grandeur, persecution, and reference. This made PARRY a valuable tool for studying the dynamics of human communication and the complexities of mental illness.
However, PARRY’s realism also raised ethical concerns, particularly regarding the potential impact on users interacting with a simulation of mental illness. Some researchers worried that prolonged exposure to PARRY’s delusional behavior could have negative effects on vulnerable individuals, though these concerns were largely mitigated by the controlled nature of the experiments.
Despite these ethical considerations, PARRY’s contributions to the field of artificial intelligence were significant. Its lifelike interactions and nuanced portrayal of paranoid schizophrenia helped pave the way for future advancements in chatbot technology.
Parry and ELIZA met once. You can read their interesting conversation here.
1980s- Racter: An attempt at Generative AI
Developed by William Chamberlain and Thomas Etter in 1983, Racter was unlike the two chatbots that had come before it, boasting a unique approach to generating text-based conversations.
Racter was designed to simulate a writer, capable of producing original prose and engaging in literary exchanges with users. Its creators envisioned Racter as an experiment in machine creativity, exploring the potential for computers to generate coherent and compelling written content.
What set Racter apart from previous chatbots was its use of a technique known as “pragmatics-driven text generation.” Instead of relying solely on pre-programmed responses or pattern-matching algorithms, Racter employed a rules-based approach to generate text based on the context of the conversation.
Users could engage with Racter in a variety of ways, from asking questions and seeking advice to simply conversing about everyday topics. Racter’s responses were often surreal and unpredictable, reflecting the experimental nature of its design.
One of the most famous examples of Racter’s output is “The Policeman’s Beard is Half Constructed,” a book composed entirely by the chatbot. Filled with nonsensical poetry, philosophical musings, and whimsical anecdotes, the book captured the imagination of readers and showcased Racter’s unique writing style.
Despite its groundbreaking approach to text generation, Racter ultimately faded into obscurity, overshadowed by more advanced AI technologies that emerged in the years that followed.
1990 to 2000 - The Rise of Smarter Bots: ALICE, SmarterChild, and Cleverbot
As technology advanced, so too did the capabilities of chatbots. In the late 1990s and early 2000s, we witnessed the emergence of ALICE, an ambitious project aimed at creating a more sophisticated conversational agent. While ALICE’s responses were often more coherent than the chatbots that came before it, they still occasionally veered into the realm of absurdity, much to the delight of users.
Meanwhile, on instant messaging platforms like AIM and MSN Messenger, a new breed of chatbots emerged. Among them was SmarterChild, a virtual assistant capable of providing weather forecasts, sports scores, and even rudimentary games. Though SmarterChild’s responses were sometimes hit-or-miss, its quirky charm endeared it to millions of users around the world.
Not to be outdone, Cleverbot entered the scene with claims of artificial intelligence and the ability to learn from its conversations. While its responses often bordered on surreal, engaging with Cleverbot became a popular pastime for curious internet denizens.
2010s - Enter Siri, Alexa, and the Age of Voice Assistants
With the advent of smartphones and smart speakers, chatbots took on new forms and capabilities.
Siri, introduced by Apple in 2011, was one of the first mainstream virtual assistants to gain widespread popularity. Siri leveraged natural language processing (NLP) to understand user queries and perform tasks such as setting reminders, sending messages, and providing information. While initially limited in functionality, Siri’s integration with Apple’s ecosystem and its playful personality endeared it to users around the world. It became a household name, providing users with answers to their questions, reminders, and even jokes (albeit corny ones).
Meanwhile, Amazon introduced Alexa. Launched in 2014, it took the concept of virtual assistants a step further by introducing hands-free voice control via smart speakers such as the Echo. Alexa’s deep integration with Amazon’s services and its ever-expanding library of skills allowed users to perform a wide range of tasks, from controlling smart home devices to ordering groceries with a simple voice command. Alexa’s ability to understand context and engage in natural, conversational interactions set a new standard for virtual assistants.
And then came Google. Google Assistant, introduced in 2016, brought Google’s expertise in search and artificial intelligence to the realm of virtual assistants. Built upon Google’s Knowledge Graph and powered by advanced machine learning algorithms, Google Assistant offered unparalleled accuracy and comprehension in understanding user queries. With its integration across Google’s suite of products and services, Google Assistant became an indispensable tool for millions of users worldwide.
As voice recognition technology improved, so too did the conversational abilities of these virtual assistants. Users could now engage in natural, voice-based interactions, making tasks like setting alarms and checking the weather as simple as speaking aloud.
2020s - ChatGPT: A new age begins
It came out of the blue for most people. But the story of ChatGPT starts with OpenAI.
Founded in December 2015, the organization was born out of a shared vision among its founders—Elon Musk, Sam Altman, Greg Brockman, Ilya Sutskever, Wojciech Zaremba, and John Schulman—to ensure that AI is developed and deployed responsibly, with a focus on safety, transparency, and ethical considerations.
OpenAI’s journey began with a bold mission: to create artificial general intelligence (AGI)—a form of AI that can understand, learn, and adapt across a wide range of tasks and domains (like a human).
Incorporated as a non-profit organization, OpenAI quickly garnered attention and support from leading figures in the tech industry and beyond. With an initial endowment of $1 billion and a team of world-class researchers and engineers, OpenAI embarked on a journey to push the boundaries of AI research and unlock new capabilities that were previously thought to be impossible.
One of OpenAI’s most notable achievements came in 2018 with the release of GPT (Generative Pre-trained Transformer), a state-of-the-art language model capable of generating human-like text based on input prompts. GPT laid the groundwork for subsequent iterations, culminating in the release of ChatGPT.
Fast forward to today, and ChatGPT and its siblings DALLE and Sora, can understand and generate just about anything you can imagine.
It truly is a new age and the journey has only just begun.
Also in the 2020s - Superseek: Custom goal-driven ChatGPT for businesses
Superseek is a platform to create custom AI chatbots trained on a business’s content – website, help docs, documents, etc.
Businesses add Superseek to their website to accelerate lead generation and instantly resolve customer support queries.
Create a custom AI chatbot for your business for free.