2018: The Rise Of The Chatbot (part two) January 26, 2018

Here we return with part two of our Chatbot series, where we are discussing the rise and application of a technology that's proving to be more than just another marketing tool or customer service automation.

Let’s dive straight back in and look at this so called ‘efficiency boost’…

What can chatbots achieve in cost savings?

Take the healthcare industry.  With tools like Alexa, Google Home and Siri already learning your food preferences, your quickest journey home (based on learned behaviours and traffic stats) and even making recommendations based on your shopping habits, it’s only to be expected that these same technologies could be applied beyond our lifestyle and into our health management.

We’re already using robotics for complex surgery, so why not use them for triaging minor ailments?

Where patients are already linked up to smart devices like FitBits, Jawbones and Apple Watch, recording heart rate, oxygen saturation and blood sugar levels, this could mean the difference between taking half the morning off work for a GP appointment and an ‘at-work’, instantaneous data transfer.

This procedure could give your virtual healthcare provider a clinical overview of your condition and perhaps even be mapped against historical submissions, creating patterns and a predictive model that could give you forewarning of future illness. Maybe not everyone's cup of tea, but a powerful notion nonetheless.

When you look at how these technologies could relieve the burden on the NHS and HSE, in administration alone, you start to understand how the potential to transform the healthcare industry is genuinely profound. Imagine what modern healthcare could achieve if budget was redirected from form-filling, colds and flus, to global-scale medical research and new treatments?

Risks and managing compliance?

Where compliance, risk and chatbots meet is an area that fascinates humankind. Being concerned with robots and technology effecting our lives has been occupying our imagination since the 60s and the Jetsons took to the screen.

Fear not. Where transaction handling and the provision of advice is concerned, there are definite, hard and fast rules. In particular, the requirement that the bot must declare itself prior to engaging with the user. The acceptance of these Terms and Conditions between the bot and the user must be explicit and unequivocal.

Naturally, this brings up all sorts of questions around simple message automation (“Disclaimer: By using this service you acknowledge and accept that any and all advice(s) are provided in part or in full by a chatbot etc etc "), or deeper, into AI and machine learning. An automated message is one thing. But leaving the declaration to the bot is another.

Does the bot know to ask because it is 'compliant by design'? Or by nature? Can the bot decide to give erroneous advice or fluff a transaction because it wants to?

By now you will have heard of Sophia the Hanson Robotics creation. If not, then look her up. She.thinks.for.herself.


The good news is we don’t expect that level of autonomy to be applied in a simple service industry chatbot, but it’s a risk that must be assessed and managed nonetheless. In much the same way as Amazon and Google will have invested millions ensuring their home-listening devices can’t and don’t develop a life of their own and process unauthorised transactions, anyone developing chatbot Apps will have to demonstrate thorough risk assessment and undertake in-depth user testing before market release.

With the GDPR deadline on the horizon , naturally compliance of any kind is something that’s never too far from our thoughts when it comes to business these days. So, what better example than one in the legal sector itself?

ExpertGuides.com reports on DoNotPay, dubbed the first ‘lawyer robot’ and a chatbot service that provides legal advice on whether or not to pay a parking ticket.

“In 26 months, DoNotPay has taken on 250,000 cases and won 160,000, giving it a success rate of 64% appealing over 4 million USD of parking tickets. After this success, the AI lawyer went on helping people with flight delay compensation, as well as acting as a guide for refugees navigating foreign legal systems.”

What’s interesting about this isn’t the success rate of the bot for dodging parking fines, but that it proved industry compliance issues, like those in France where the offering of legal advice is heavily regulated, can be circumvented by the qualification and knowledge base of the bot itself.

If the bot is programmed to analyse cold, hard legal fact and advise based on that, without the grey areas only injected by human intervention, then compliance becomes less of an issue and more of a tick-box exercise.

So there you have it. Chatbots are big news for 2018. Interesting to think they have been ticking away in the background of our lives for some time now, but all of a sudden they are taking off. It just goes to show that it takes the right conditions for an idea to really catch on. That’s the interesting thing about digital transformation on a global level, it’s not uniform. In some areas we are light years ahead of ourselves and in others we are still holding back, treading carefully until we know more…or until we accept the inevitability of progress.

Satya Nadella himself has said that chatbots are an inevitability. Back in 2016 at the Microsoft World Partner Conference, the Microsoft’s CEO delivered that ‘chatbots are the new Apps’ and his prediction that they will ‘fundamentally revolutionise how computing is experienced by everybody.’

Watch this space. 

Miss part one? Catch it here.

Interested in how chatbot technology could grow your business? Get in touch! Drop us an email below or call us on 028 90 87 2222

 Security code


Etain Ltd make custom software; we aren’t legal professionals. The information in this article is presented in good faith, and is intended as a high-level guide to help you understand the technologies discussed.

Return to blog

We're Hiring

Interested and want to know more?Send us an email