Are chatbots really intelligent?

Are AI chatbots really intelligent?

The first thought that we get, as we come across a chatbot is that, if they are really “intelligent”.  And if they are, then how? Is it some sort of Artificial Intelligence? Prior to entering this field, I thought that the chatbots in retail are just fed with numerous Question and Answer pairs(FAQs). In reality, a ChatterBot seems to be a box of surprises.

What makes a conversational AI chatbot intelligent?

A friend of mine said, “chatbots remind me of kids”. I thought it might be an interesting example to explain how chatbots work. Remember how our parents taught us to say thank you? The picture illustrates 3 scenarios, which are kind of self explanatory.

How chatbots learn

What is interesting here is the 3rd scenario, where nobody is teaching the kid to say thank you, but she still does. 

Here we should notice 3 things:

  1. The child understands from its previous experiences that the 3rd scenario is also about giving and taking. 

  2. The child understands it has received something pleasing.

  3. The child knows it should thank the other person every time it receives something from them. 

How did the child learn this? Here’s where human intelligence plays its role. Similarly, when a machine tries to do the same, it needs some intelligence too – which is nothing but Artificial Intelligence.

Can a conversational AI bot learn?

A conversational AI chatbot just cannot reply to whatever we ask, unless it knows and understands the language we speak. Knowing the language helps the bot to understand the semantics of textual data. The concept of language models help us represent any textual data from a language in the form of numbers, preserving all the semantics! So, does that mean a chatbot can speak any language if we create a language model for that particular language? Indeed!

So now that we know that our ai chatbot can learn our language, it now needs to  –

  1. Understand the intention of the user

  2. Identify any specific entities mentioned in the query

  3. Keep track of the context in which the conversation is happening

  4. Generate a reply to be sent back to the user

Relate to the kid’s example above? The way the kid’s mom taught her, our ai bot too needs someone to teach them. Basically, we need to teach our bot how to perform the above tasks. In technical terms, we call this Machine Learning

Do you see how all these fancy terms come together to form a chatbot? Sweet! 

So, can a AI chatbot answer anything and everything?

Well, depends on how you train them. Usually chatbots are trained to serve a specific use case. It may be for example a customer service chatbot, an ai assistant for retail shopping or could be even industry specific such as chatbots in retail or chatbots in fashion industry.  These domain specific bots are more powerful due to their specificity and hence are more popular as well. However, generic chatbots can also be built which can address more general queries which could be as vague as “what should I do today to stay positive”. These chatbots require a huge amount of data and are often the hardest to train. 


  1. Chatbots are not simply hardcoded machines but use Artificial Intelligence and deep learning techniques to learn from given data to try and imitate humans while conversing.

  2. Chatbots can be trained to serve any specific AI use case or to carry general human conversations depending on the need and specificity.

Do have a look at “Are AI and Machine learning the same” by Tejas Venugopal to get a clear picture of AI, Machine Learning and Deep Learning.  

Also i recommend the article by Joseph Mathew Traditional Online Chatbot or Domain Specific Conversational AI? What can create a delightful customer experience online? to know why domain specific chatbots could be a better choice for your business.

Curious Much? Drop us a note at or visit our website and let’s start a conversation

Are AI and Machine Learning the same?

Are AI and Machine Learning the same?

We have all heard or read about how AI is the next big thing. It seems like every 3rd article or blog is about AI and how companies are leveraging this shiny new tech to solve some of their biggest problems. And the same words are seen together. AI vs ML, Machine learning vs Deep Learning, Neural Networks ….. and so on.

Every so often I get asked (by engineers and non-engineers both) “Are AI and machine learning the same?” So let’s answer this.

What is AI? What are its goals?

I don’t want to give a textbook answer here on the AI meaning. I am sick of them myself. So here it goes…

AI’s primary goal is to create an artificial intelligence. This may or may not include a robotic body as such. That’s only for movies, not in real world always. An artificial intelligence needs to be capable of most, if not all, the things that humans do.

  1. It should be able to learn by observation or actions.
  2. It should be able to decide on a goal
  3. It should be able to create a plan to achieve this goal in an optimal way.
  4. It should be able to sense or feel and have an awareness of its surroundings
  5. It should be able to reason and draw conclusions and so on..

As you can see this is a pretty lofty goal. It is quite abstract as to how this can be achieved. And that’s the main point. AI is abstract in general. It has goals but needs ways of achieving them.

How does machine learning assist AI goals?

Machine learning is one of many tools that is used by an AI to achieve its goals. Machine learning unlike AI is very clear about its goals. Machine learning wants to create function approximations or models for some input and output combinations (we call this “training data”). Using these AI models, it can later generate outputs (we call this “predictions”) on unseen data. To create these models, it uses various AI algorithms which have hyper-parameters to help tune them.A very easy example is,

Input (X)         Output (Y)

1                      3

2                      6

3                      9

4                      12

It’s clear from these combinations that the output is 3 times the input. So the function would be,

Y = 3X

Now this becomes our model. If we have unseen X values, we apply the model to predict the Y values. But of course, the AI algorithm is never this simple in real life. Here X could refer to a vector generated by text, image, audio, etc and Y could be any useful prediction.

These types of predictive models are used by AI to reach its goals.

Where exactly is Machine Learning used?

So we know that machine learning is used to create predictive models. But where exactly are these models used? What can you do with them? Let’s look at some relevant examples for businesses.

  1. Personalized recommendations – When netflix recommends movies based on your previous actions, when amazon recommends products for you to checkout, Machine learning is used to predict them
  2. Search engines – Google uses machine learning to provide you with search results for your queries
  3. Maps – Google uses machine learning to recommend the best route for your journey based to driving time calculations taking into consideration the live traffic and road closures data
  4. Virtual Assistants and Customer Service Chatbot – Natural language processing techniques of machine learning are used to provide a conversational AI experience to the user. This can be used in a wide variety of fields like customer service and customer engagement in ecommerce.
  5. Video Surveillance uses machine learning to reduce human effort by predicting any unusual behavior in video feeds
  6. Face Recognition – Many smartphone providers use machine learning to help you unlock your phone with your face
  7. Spam filtering – Email and SMS providers use machine learning to help filter out unwanted messages

AI vs Machine learning vs Deep learning

This is another familiar question – How is deep learning different from machine learning? If I had a nickel…

Machine learning’s goal is to create predictive models. Deep learning wants to do the same. Really. They’re like twins.

While machine learning likes to use traditional models like SVM, Decision trees, etc Deep learning uses fancier (read as “more complex”) neural network AI algorithms which work like the human brain (in theory). And that’s it. Everything else is pretty much the same.

I usually don’t refer to Deep Learning as a separate term. To me they are the same. Deep Learning is Machine Learning with updated AI Algorithms.


To summarise,

  1. AI vs ML – AI is an abstract concept with many goals. Machine Learning is a specific tool used by AI to realise them.

  2. Deep Learning is Machine Learning with updated algorithms.

For more content on AI and how businesses can leverage it, please check this article from Dinesh Sharma  “Can AI make your business far more intelligent

Curious Much? Drop us a note at or visit our website and let’s start a conversation!

How AI chatbots will transform the retail industry in 2020?

How AI chatbots will transform the retail industry in 2020?

Consumer Good manufacturers are an integral part of the retail value chain and while a lot has been talked about application of AI and Chatbots in retail, this article focuses specifically on how Consumer Good brands too can benefit from chatbot technology. It is a known fact that relationship between brand manufacturers and retailers have always been strained through the decades and the quest for the last-mile control for consumers can be seen with more and more brands adopting a Direct-to-Consumer strategy in recent years. A great example of this being Nike announcing that it is pulling back from Amazon in November 2019 after 2 years of signing the partnership deal.

So can AI chatbots help Consumer Product companies? And if so How?

When I speak to consumer good companies about AI chatbots and their potential to impact P&L, one of the most common objections I get is that “we do not sell online but instead we sell through ecommerce channel partners. So how is AI bot relevant for us?”. Having helped some of the largest global CPG companies implement AI assistants or chatbots over last 3 years, I assure you there is a lot chatbots in retail can offer even if you are not selling online. Here are the top 4 chatbot use cases that you can look to implement in 2020

Top 4 chatbot use cases that you can look to implement in 2020

Top1: An AI Chatbot on the brand’s website helping consumers with their product research

Regardless whether you sell online or not, you still have a website and you still have a steady flow of consumers visiting your website every month. Right? (If the answer to this question is “no” then skip this use case – this is not for you). Your consumers are flocking your website for just one reason. They want to research more about your products, they have questions and are searching for answers on your website and this is your chance to engage them, give them what they need and help them decide. An AI chatbot that has been trained well on your products and knows everything about your brand is a great tool to deploy on your website, as a companion app, making it super easy and convenient for your consumers to find the information they need and also personalize the interaction based on the consumer’s context. With every conversation, it is not just that your AI is learning and getting better, but you are also collecting rich consumer behavior data which, if harnessed correctly, can produce deep and unique precision marketing insights on topics ranging from online experience to new product ideas.

Chatbot on website
Chatbot on Dulux website

Top2: An AI chatbot on select product pages of your ecommerce channel partner’s website

Most of the global consumer product manufacturers sell online via e-commerce channel partners and even if they have a direct-to-consumer strategy, it is likely that a bulk of the online revenue is still coming from e-commerce channel partners. If you are one such brand and if you are thinking how can an AI chatbot help you drives more sales, why not explore with your e-commerce channel partner the possibility of having your AI assistant on your product pages of their website? An AI chatbot trained about your products that can answer questions of consumers instantly is a great tool to differentiate your brand from your competition selling on the same channel partner’s website. But it does not stop just there. Remember the conversation data still comes to you (it is your AI chatbot after all and all interactions with the ai bot gets saved) and your marketers get to benefit from all the insights hidden in these conversations.

Top3: An AI chatbot available through a simple QR Code on your retail partner’s store

Is chatbot technology only for the online world? No, they are not and infact they can be your differentiator even in physical stores. Your consumers are buying your products from retail stores and you are left at the mercy of the store associate to sell your products. Why not have your chatbot made available through a branded QR Code inside the store? Your consumer has a question on your product, she scans the QR Code and your AI assistant pops up in her mobile phone assisting her with her questions, connecting her to your in-house experts seamlessly and also probably offering her a personalised discount voucher instantly. Wouldn’t that disrupt your competition? One such great example is Dulux paints, the No1 paints brand in UK that launched their AI assistant in Homebase stores in UK. (If you are based in London and you want to see first-hand how this works, you should visit one of the Homebase stores and check it out yourself.)

Top4: An AI chatbot that converts banners into conversations using Conversational Ads

Digital Ads in the form of intelligent conversational interfaces gives your brand marketers a new and powerful way to engage your consumers and drive productive outcomes. Banner ads are one directional and non-interactive while conversational Ads are two way interactions that can be personalised to the consumer’s context at scale. Also, imagine the effort and cost this saves you from creating a maze of landing pages, forms and navigation steps on your website prior to launching your Ad campaign only to get people come on your website. With conversational Ads, you engage the right customers instantly and drive meaningful outcomes.

In summary, even if consumers do not buy products from the manufacturer’s websites directly, consumers still expect their favourite brands to serve them and help them make the right choice. AI chatbots offer consumer product manufacturers a whole new opportunity to engage their consumers and differentiate from competition and here are the top 4 ai chatbot use cases   for CPG companies

  • An AI Chatbot on the brand’s website helping consumers with their product research.

  • An AI chatbot that runs on product pages of the e-commerce channel partner’s website helping consumers with their product choice on behalf of the manufacturer bringing the insights back to the brand.

  • An AI chatbot available through a simple QR Code in the retail partner’s store to engage in-store consumers and answer her questions.

  • An AI chatbot that converts banners into conversations using Conversational Ads.