What is a chatbot?
Chatbots and voicebots, a new way of serving customers
In recent years artificial intelligence has completely redefined the way businesses and customers interact, driving the emergence of chatbots and voicebots capable of handling enquiries, resolving issues and supporting sales processes with a level of personalisation that was previously unthinkable. The advent of agentic AI has accelerated this evolution, transforming these systems into true digital agents that not only converse, but also understand context, reason, act and learn within controlled frameworks, collaborating with human teams and executing end-to-end business tasks.
Update: March 2026
Reading time: 13 minutes
Process and Digital Automation Manager at Iberdrola
Innovation and technology management for Iberdrola’s Customer Service
Uses and applications of chatbots
A chatbot is software based on Artificial Intelligence that conducts real-time conversations via text or voice. They are not magic, but science. They use advanced Natural Language Processing (NLP) tools, multimodal base models and machine learning capabilities to understand the user's intent, interpret the context and provide useful responses or carry out actions.
Today we distinguish between two well-established categories:
- Chatbots (text): found on websites, apps, social networks, WhatsApp and other digital channels.
- Voicebots (voice): assistants that converse via telephone, conversational IVR, smart speakers or internal voice-based tools.
Both share natural language understanding (NLU) technologies, workflow orchestration and predictive models, enabling a smooth, frictionless experience.
The main recent transformation is the incorporation of agentic AI, capable of designing its own action plan, deciding which corporate tools to use, executing end-to-end tasks and learning from the results to improve efficiency.
This turns chatbots and voicebots into operational agents that no longer merely respond, but act: they verify identities, query internal systems, generate content, complete forms, manage payments or assist human teams with complex processes.
Types of chatbots
Depending on how these tools are used, chatbots can be divided into six types:
Linear conversational chat
These are based on a decision-tree architecture and are not very smart. Their flow of answers is determined by a linear chain of stages, so they give automatic responses rather than establishing a fluid conversation.
Non-linear conversational chat
Thanks to Machine Learning and NLP, they can interpret the user's intentions and the context of the conversation in order to respond accurately. If a user makes a request, the bot will understand it and return a few options tailored to his or her requirement.
Hybrid conversational chat
A combination of both the above that enables fluid and personalised conversation with users. If the chatbot doesn't know how to answer a question, it immediately notifies a human agent to add the answer into its database.
Voice chat or voicebot
There are chatbots that work with voice commands, such as Amazon's Alexa and Apple's Siri. They are simpler to use but can be more effective than other types of bots. For example, they can translate into another language in real time.
AI chat
Artificial intelligence chats simulate human conversations. This is made possible by natural language processing (NLP), base models and machine learning, which allow them to listen, process and understand our expressions to respond as naturally and as personably as possible.
Generative AI chat
Generative AI chatbots are similar to conventional AI chatbots, but go one step further. While AI chatbots focus on understanding language and generating responses that are as human as possible, generative AI is capable of creating content, such as images or videos. Deepfakes or chatbots such as OpenAI's ChatGPT are examples of this type of AI, which can generate images, stories, songs and more.
Intelligent agents (agent-based AI)
These are autonomous systems capable of understanding their environment, reasoning, planning and executing actions, working iteratively until a goal is achieved. These agents not only interpret instructions, but also make decisions, break down complex tasks into steps, coordinate subtasks and communicate with one another within an orchestrated business workflow.
Unlike traditional deterministic workflows, intelligent agents integrate language models (LLMs) as a reasoning engine, enabling them to respond flexibly to unforeseen situations and maintain fluid conversations. Each agent may have its own objective, a set of assigned tools and information sources – for example, access to databases, corporate APIs or internal systems – and delegates to other agents once its task is complete.
The ultimate virtual assistant
Although they seem like a recent invention, chatbots go back to the 1960s.Eliza was the first. This was a rudimentary software program created in the Artificial Intelligence (AI) laboratory at MIT. It could simulate a conversation by using a pattern comparison methodology. One thing that is certain is that the chat was more monologue than conversation.
Despite its limited functionality, Eliza fired the starting gun for a dizzying race that has led to today's chatbots, which are far more intelligent and sophisticated. This qualitative leap has been so great that many companies are adopting chatbots as a new, fast, efficient and profitable way to provide customer care services.
Today 62% of consumers would rather interact with an AI chatbot than a human and by 2030, with the upward trend of chatbots, this market is expected to be worth $27.3 B.
That is why from the 1960s, when Eliza ushered in this revolution, to today's agents based on large multimodal models, the evolution has been extraordinary. Now chatbots and voicebots are not just part of the future: they are already a strategic component of customer service and internal business operations.
And the progress continues. Agentic AI will open even more doors in automation, personalisation and efficiency, taking the conversational experience to a whole new level.
Steps required to implement a chatbot
Attending to the user experience
Give it a name, define its personality and use natural language to avoid frustrating the customer.
Analyse the customer
Define the type of business that will be served by the bot and get to know the user to determine whether their profile fits with this technology.
Define the type of chatbot
The most-suitable bot will be determined depending on its purpose —product sales, customer support, content generation—.
Update periodically
It is important to analyse their performance and change conversational flows regularly to optimise their performance.
Set targets
Does it improve customer relations? Does it open new sales platforms? Can it connect with the new generations? Depending on the answers, one or the other will be the better option.
Source: Gus Chat.
Advantages and disadvantages of an enterprise chatbot
The use of chatbots, is becoming more and more common and is gradually becoming a part of our daily lives, especially in the field of customer service, although not only in this field. Nowadays, virtual assistants are beginning to be used in companies to promote sales, retain customers and generate content of interest.
Chatbots have numerous advantages: 24/7 365 days a year availability, fast and efficient response, constant learning, cost savings, easy personalisation and even social media management! And the disadvantages? Although natural language processing is getting better, these tools don't capture certain characteristics of human language such as double meanings, sarcasm and moods, which can lead to customers getting frustrated. Also, many people, especially the elderly, reject these types of technological advances and there is the possibility that the system may fail mid-conversation, causing discomfort and mistrust in the user.
Differences between virtual agents and chatbots
The terms "chatbot" and "virtual assistant" are often confused, mainly because they are very similar in essence. However, there is a subtle difference between the two.
AI chatbots process language and generate responses that mimic human speech. If you ask an AI chatbot: "Where’s the Sagrada Familia?", it’s likely to answer that it’s in Barcelona, Spain, and to tell you a bit about the history of the cathedral.
A virtual assistant, however, combines AI language processing and intelligent search with robotic process automation (RPA). Its goal is to act on the user's request to minimise human intervention. So, if you ask the same question: "Where’s the Sagrada Familia?", it will not only tell you that it’s in Barcelona, Spain, but, depending on your location, it might send you a link with maps, flights or different options to get there from wherever you are.
In the current context many corporate projects are moving towards agent-based models where the boundary between the two concepts is becoming even more blurred.
Iberdrola Group’s current conversational ecosystem
At Iberdrola we have an ecosystem of chatbots and voicebots that deliver real value across several countries. In Spain, chatbots provide extensive support for customer service, including solutions for payments, enquiries via the Customer App, billing, sales and information regarding the Bono Social (Social Bond), as well as specific initiatives such as the win-back chatbot in Portugal and internal tools for agents that optimise day-to-day operations. In the voice domain, key projects include the Bono Social voicebots, sales and contract verification systems and other informational voicebots.
In the UK (ScottishPower) advanced solutions such as Bruce Assistant are in operation, handling both calls and emails, alongside bots for automated replies, debt recovery and outbound calls. Furthermore, Koiné (ITNow) functions as a 24/7 internal IT support chatbot, handling SAP account unlocks, password resets and frequently asked questions, accessible via the corporate portal, thereby consolidating technological self-service for employees.
In Brazil (Neoenergia) a transactional bot on WhatsApp provides mass service to distribution company customers whilst a digital voice agent automates debt negotiation through integration with the negotiation portal and APIs, reducing friction and processing times.
As can be seen, this entire suite forms a consolidated portfolio of conversational automation that is already in stable operation in Spain, Brazil and the UK, accelerating processes, reducing processing times and improving the customer experience across multiple channels. Furthermore, thanks to the success of the previous bots, work is underway to expand the scope to include informational chatbots for the General Shareholders' Meeting and even AI-based public top-up services.
Related content
What is Artificial Intelligence?
Although still at a very early stage, is laying the foundation to lead a revolution comparable to that generated by the Internet.
Internet of Things (IoT)
Objects will become smart and the world will undergo a fourth industrial revolution.
'Smart cities': The digital transformation of the cities
We present the world's smartest cities.
The self-driving car
The self-driving car developed by Google is called Waymo, and doesn't need the Internet to travel safely on the road.