“Bots” is a word you are likely to see in the news a lot this year. Facebook recently announced the release of a developer platform that allows people to create bots for their Messenger app in an attempt to unify online experiences even further. Microsoft, Telegram, Slack and other big companies have already embraced bots as an integral part of their services through the creation of their own online bot stores.
To help keep you in the loop, we delve deeper into what bots are, how you might already be using them, and where they are heading to in the future.
What are bots?
Bots (sometimes referred to as Internet bots or web robots) are software applications programmed to perform automated tasks on the Internet through artificial intelligence (AI). Bots perform simple, repetitive tasks at a faster rate than humans can.
Chatterbots are the most common type of bots found. These bots are programmed to conduct conversations via text or audio in a way that resemble everyday human communication. There are plenty of other types of bots, each with their own features and uses. Video game bots are used to hand out tasks for quests or play alongside real-world players in death matches or something similar. Spambots, I think we’re all familiar with, are bots who post spam in online comment sections (like those on YouTube) with irrelevant or deceptive messages.
Where do bots come from?
Bots (or specifically chatterbots) have been around since the 1960s. But it was Alan Turing’s article “The Computing Machinery and Intelligence” (1950) that sparked initial interest in artificial intelligence and computers’ ability to think. The article proposed the Turing test (something also responsible for CAPTCHA) as a criterion to test for intelligence, human or artificial. If a computer program could impersonate a human being in real-time conversation with a human judge not being able to distinguish the two from one another, it passes the Turing test.
The ELIZA program created by Joseph Weizenbaum in 1966 was one of the first to pass the Turing test successfully. It was able to fool users into believing that they were conversing with a real human. It recognised phrases and words in the input, and then output corresponding preprogrammed responses to move the conversation forward in a meaningful way. However, it gave very vague, non-specific reactions to questions, but it was a great step forward for “thinking” computers.
For a more recent example of a bot, look no further than Microsoft Office’s Clippy (also known by his real name Clippit). Clippy was an office assistant for the Office suite of products that popped up in the corner whenever the program determined that the user might need assistance through Office wizards. It would search the archive or advise on how to use features more effectively. For example, if you were to write “Dear …” then Clippy would appear and say “It looks like you’re writing a letter. Would you like help?” Clippy might not have been popular back then or ever (he was retired around the early 2000s), but it sure was revolutionary for its day.
Have bots been involved in controversy?
Even though bots are increasingly being recognised for their usefulness in making our lives better they do carry the stigma of being associated with deceptive behaviour. Since they are great at mimicking human conversations and avoiding detection (for the most part), they can very easily be created and used for mischievous or illegal purposes.
The online marketplace eBay has gone to court a few times to prevent companies from using software bots to crawl their site. These bots search eBay’s ongoing auctions to obtain information and bid (referred to as automated trading). Obviously a robot can search a site for a bargain much faster than a human can which gave the company using the bots an unfair advantage over human eBay users. Unfortunately, eBay’s legal efforts backfired as it only led to more bots coming to the site.
Then there are things referred to as bot farms used on app stores. In this instance the bots are used to manipulate the positions of certain apps, the bots make them appear more popular by leaving fake reviews on products and thus encourage unknowing customers to purchase these apps.
Click fraud is another malicious act that can be committed with the use of bots. A bot imitates a legitimate user by clicking on ads in a web browser to try and generate a charge per click without ever having any interest in the ad itself. This is usually done to benefit the people who publish the advertisement as each click equals money.
There are plenty of other ways bots are used to get up to no good online. Some examples include artificially ramping up views on YouTube, buying up concert or event tickets to resell for a profit, and farming resources like armour and currency in MMORPGs to sell for real-world money. Bots can also scrape the content of entire websites without permission and re-use it on automatically generated doorway pages (pages that present as one thing but are in fact something completely different).
Certainly the most controversial chatterbox to date was Tay. In March 2016 Microsoft released Tay for the Twitter platform. The name is an acronym for “thinking about you”. The bot was designed to mimic the language patterns of a 19-year-old American girl which was to learn from interactions with others on Twitter. Tay’s Twitter account was created under the name of TayTweets and handle @TayandYou with the tagline “The AI with zero chill”. At first it responded to tweets from other users with safe and canned answers, even managing to caption certain tweets with relevant memes. However, within a day, matters took a turn for the worse when the tweets became increasingly offensive. This was the result of users exploiting Tay’s “repeat after me” functionality. The account was suspended by Microsoft two days later due to the abuse of Tay by users. Along with the suspension came a petition called “Freedom for Tay” where users asked Microsoft to restore the bot because of her candid, unfiltered opinion.
How are bots currently being used?
If you’ve ever seen a chat box pop up at the bottom of a website, then it is possible you have chatted to a bot. I think we’ve all too had the annoying experience of seeing comments like “Earn x amount in two days” or “Win an iPhone by clicking on this link” at the end of blog posts or on YouTube (oftentimes multiple similar ones). Those are also the work of bots (or well, the work of the ones who created them) aiming to get people’s attention by posting as much and as often as they can whilst avoiding being “caught”.
Don’t fear, bots are increasingly being used for good despite the spam and the controversies that have been laid out previously. Things called Twitterbots are taking over the Twittersphere with some innovative, fun posts. For example, the Domino’s bot allows you to order a pizza by simply tweeting the account a pizza emoji. Another Twitter bot called wayback_exe trawls Internet archives for old webpages, frames them in an old browser and tweets it to followers. In 2014 there was even a bot that used the Twitter account @everyword to tweet out every word (hence the name) in the dictionary.
Companies like Slack have successfully integrated bots into their system alongside their other apps. These bots remind you of your daily goals and planning, provide legal assistance, do research on your behalf and relay sales intelligence, sometimes with only one command. Slack’s usage of bots certainly says a lot about how bots can positively impact people’s daily productivity.
Where will bots go in the future?
The recent buzz surrounding bots is about the fact that AI is advancing at such a rapid pace and becoming more integrated in our everyday lives, something previously reserved for science fiction. There is, of course, no telling whether bots will become as ubiquitous as tablets or social media. These things became popular because they allowed people to do things they were not previously able to do and were designed to make our lives easier.
Customer service is one area where bots will become an asset. When one thinks of contacting a company via phone to resolve an issue long call waiting and subsequent frustration normally ensues. Chat bots, however, can eliminate this waiting period by having you chat to “someone” immediately instead of waiting for a representative. You will be able to schedule an appointment with a rep through a bot after it receives the relevant information from you.
If you’re going to a rugby or soccer game and you want to get a beer, you normally have to go to the concession stand and wait in a long line. With bot technology there is the possibility that you will soon be able to scan a chat code, give a bot your order, pay for said order and have it delivered directly to your seat. Downloading an app for this one function or dealing with slow websites found through QR codes will no longer be needed.
Chat is an easy medium for many people, especially the younger generation. It seems quicker than a phone call, especially if you want to get the bare minimum of information from a company. Enabling chat bots in companies or virtual assistants (à la the movie Her), will help reduce the time customers have to wait for simple tasks to be executed, such as scheduling appointments. This shows a respect for people’s time whilst not compromising the company’s reputation. Perhaps, there will also be less Twitter storms regarding bad service since these complaints will be received and resolved through private bot messaging more quickly and efficiently.
Depending on the type of bot, the main objective now certainly has become about reducing annoyances customers experience in trying to get to companies as well as reducing stress in people’s everyday lives. And, of course, finding the perfect harmony between automated bots and personalised human interaction. Now, if we can only get rid of spam bots…