Artificial Intelligence (or AI) is a bit of a buzz word these days. Although the subject of AI is very broad, for the purpose of this post I will just touch on a couple of familiar examples that have developed online in particular.

Chatbots

Chatbots are probably the most talked about area of AI at the moment and are popping up across a lot of social media and messaging applications.

So what is a chatbot?

The basic idea behind chatbots is that they mimic the conversation that you are most likely to have with a real human and thus automate the process that old style phone customer support / custom service provided.

Think of those annoying calls where you sat in a queue and were reminded that ‘ Your call is important to us’.

With a chatbot, the whole question / answer process could be automated via their Facebook messenger, getting you the information you need, in no time at all.

That’s the concept in theory anyway.

Take Dominos pizza as an example. They are pushing the concept that you can, order anyware (sic), from their common social media apps e.g. Twitter, Facebook, etc.

Taco Bell are another company that are doing something similar via Slack – check out their TacoBot.

As side from the food industry, the e-commerce main players, such as Amazon, are now investing heavily in developing new bots that will help your shopping experience.

Check out their AWS service, called LEX – it lets you build your own conversation bots.

So that’s it… the bots are in charge?

Not exactly. As good as the technology is becoming, chat bots are still far from being a perfect science.

The way they work is basically to learn the most common responses to give based on a bunch of data that it is thrown at them.

The more data they consume, the more they learn.

This concept falls short when the data that they are consuming is not the correct data.

The most recent high profile example of this was Microsoft Tay Twitter Bot. The bot was meant to learn from the people it chatted with on Twitter but bold users quickly learned they could corrupt it’s mind – it became a racist in less than 24 hours.

One major flaw with the current range of chatbots is that they are not good at understanding human emotions.

The guys over at Intercom have a nice article where they suggest that we shouldn’t be striving for a end result where people can’t tell the difference between chatbots and humans.

They feel that the bots would be great for answering simple, repetitive questions but if you need something trickier answered, then you should be routed to a real person.

They probably sum it up best by saying:

Chatbots are best left to handling computation. Things that require empathy or emotion, on the other hand, are still better handled by a human.

Check out their blog article and podcast for their take on AI trends for 2017.

AI in website design

Wouldn’t it be great if your website could design and build itself? Well obviously not so great for our line of business… but there are some companies that are attempting to go down that route.

TheGrid.io is one such company that promises that. The have an AI engine that will build your site design based on the content.

Initially there was a lot of people that signed up for their beta release but they were over a year late on launching the product which left a lot of people very unhappy.

I’ve had a look at some sites that were designed by the grid and honestly it doesn’t look like they will be creating a web revolution anytime soon.

Wix is another company that is moving into the AI design (or Artificial Design Intelligence) industry.

Wix has been around for a good while now and was used by a lot of people looking to get a site up and running fast. Their design options were pretty limited when they first launched but it looks like they have their foot on the gas when it comes to branching into the AI design area.

As with anything that isn’t bespoke, there are always compromises.

AI in SEO & Digital Marketing

This is a huge area so I’m not going to go into too much detail but AI has already replaced the traditional online advertising and search methods.

Back in November 2015, Google announced that they will veer away from using algorithms that were hand-coded by developers and start using machine-learning AI to determine a sites SERP (Search engine results page) value. Check out RankBrain for more details about how this works.

In online advertising, AI is now being used to to help companies achieve more accurate and effective ad placement.

Traditional, marketers would measure clicks and conversions for their search or display ads and use customer actions to determine success.

With the help of AI and machine learning, the likes of Google and Facebook, are optimising our campaigns to help us get the most conversions out of our ads, without relying so much on guesswork.

The future

AI has come along way from the Microsoft Clippy days (remember him!)… but it is definitely going to be integrated in a lot more of the web over the next few years.

As with all new technologies, I don’t think it will be the end of the designer and developer but will rather gives us new tools to work with, allowing us to create more creative and innovation websites.

Tell us what you think in the comments section below.