The original source of automated dialogue systems like social bots or chatbots dates back as early as the 1960s. In 1966, the computer scientist Joseph Weizenbaum developed ELIZA, a computer-based language-processing system, which is considered an ingenious early application of artificial intelligence. Thanks to huge technological leaps made over recent years, there have been significant advances in data processing and consequently in the development of artificial intelligence. The hurdles that stood in the way of bots, which are small mini-robots with their own online identity, being developed and used have now therefore largely disappeared. These mini-robots can be programmed with comparatively little work and expense.

The term "bot" is a contraction of the term "robot". Bots are computer programs that, once activated, work automatically online without any human input. These automated robots are used in a variety of ways. Wherever bots can be used to answer or even comment by themselves, they are now being used to do so. They can sometimes be difficult for people to spot, because the programs interact with their digital conversation partner via their own online profile almost exactly like a real person. This means that these mini-robots can be used for sometimes dubious purposes, depending on their objective. Depending on the bot's application, they are referred to as social bots or chatbots.

Bots on social networks: Social bots

Social bots work via social network accounts. These are used, for example, to share posts, links or to comment. In principle, social bots can be separated into "good" and "bad" social bots. Good social bots can be used, for example, in order to automatically publish tweets on Twitter or posts on Facebook. Bad social bots could, for example, be used to spread huge volumes of links to malicious websites.

In addition, bad social bots often use "fake accounts". These fake accounts or social bot profiles are so deceptively realistic that it can be difficult for internet users to tell them apart from real people. This can be a particular problem when it comes to intentional manipulation of public opinion online through "fake news". There is no doubt that there are social bots acting as troublemakers on the internet and that they could potentially be used for cyber types of attack, not just to influence opinions. Individuals could also be the direct victims of such attacks. A bot could, for example, be adjusted to match the identity of a person and post comments or send messages as them with links to malicious websites. Of course, behind every bot is the intention of the programmer, which means that in principle the mini-robots can only do what they have been instructed to do. In any event, internet users should adopt a healthy scepticism towards comments made online, and where applicable, check the information in verified sources. If you see unusual posts made by a friend or an acquaintance, you should speak to them in person and clarify what is going on. It may well be that their profile has been hacked.

Bots in messenger services: Chatbots

Chatbots are deployed wherever they can be used to automatically answer questions and comment. Companies especially are increasingly using chatbots to answer customer questions directly in a chat window. Even news portals are doing the same, like the German site Tageschau, which has its own chatbot named Novi-Bot, which users can subscribe to in order to chat with Novi about topics currently in the news. In economic settings, the technology is generally seen as a value-added to help automate repetitive processes.

Of course, the chatbot technology can, however, be misused for the purposes of internet criminals. For example, a chatbot using a stolen profile could send contact requests in order to gain the trust of the victim. Once the contact request is accepted, the chatbot might, for example, seek out personal data. Users of messengers should therefore check carefully which contact requests they accept and which they would be better off rejecting. Stay on the safe side and ask, if applicable, your friend in person whether they sent you a contact request via this service.