chatbots

Chatbots auf dem Vormarsch

85 Prozent der Kundenservice-Anfragen werden bis 2020 von Chatbots geklärt werden

 

von Jan Gerkens

Künstliche Intelligenz (KI) mit all ihren Forschungsfeldern ist ein Bereich, der nicht an Fahrt zu verlieren scheint. Die Möglichkeiten, KI einzusetzen, reichen von autonom fahrenden Fahrzeugen bis hin zur Krebsfrüherkennung. Für Unternehmen wie Facebook oder Google sind insbesondere die sprachgesteuerten Systeme und Chatbots interessant, die Künstliche Intelligenz ermöglicht.

In einer aktuellen Studie von Gartner, einem Marktforschungs- und Analyse-Unternehmen, wurden Prognosen getroffen, die uns überraschen aber auch neugierig stimmen. Demnach sollen Chatbots im Jahr 2020 bereits an 85 Prozent aller Interaktionen zwischen Unternehmen und Kunden beteiligt sein. Zwar sind jene Interaktionen noch sehr simpel, sollen aber Kunden-Wartezeiten auf ein Minimum reduzieren und Mitarbeitern langfristig die Möglichkeit geben, sich komplexeren Aufgaben zu widmen. Die Studie geht sogar noch einen Schritt weiter und prognostiziert, dass eine Durchschnittsperson in Zukunft mehr Unterhaltungen mit einem Chatbot führen wird, als mit ihrem Ehepartner.

Derweil arbeitet das Forschungsteam von Facebook daran, die Qualität der Unterhaltungen zu verbessern und den Chatbots eine gewisse Menschlichkeit zu verleihen.  Ein wichtiger Zwischenschritt hierbei ist es, den Programmen beizubringen, verhandeln zu können. In vergangenen Tests sollten Chatbots beispielsweise erörtern, wie sie verschiedene Gegenstände untereinander aufteilen. Im Rahmen eines Dialogs lernten die Chatbots hierbei ihren gegenüberstehenden Chatbot kennen, um seine Präferenzen zu erfahren. Das Ergebnis war, dass sich die Chatbots auf Deals einigten und die Gegenstände nach Vorlieben aufteilten.

Eine Eigenschaft, von der wir uns noch einiges abschauen können.

 

Quellen:
https://www.gartner.com/smarterwithgartner/gartner-predicts-a-virtual-world-of-exponential-change/
http://www.handelsblatt.com/unternehmen/it-medien/chatbot-projekt-von-facebook-kuenstliche-intelligenz-ausser-kontrolle/20133670.html

We need to talk… about automation

It’s time for us humans to do some deeper machine learning

VORN Automation

Von Tom Davis

What percentage of the manufacturing process of your smartphone is made by man versus machine? What percentage of the production & distribution of the food that ends up on your plate comes from man versus machine? Most of us will assume a high percentage of machine involvement in the manufacturing of our phones and a low percentage in the production of our food, but that could very well soon change. The question is, should we know how man-made each of the products & services we use are? Does it even matter?

 

The rise of the machines

There are great advancements being made with robotics and machine learning leading to artificial intelligence that is transforming the way we do business from manufacturing to distribution to marketing.

For example, there have been a series of truck platooning tests carried out in the last 12 months within the EU. Truck platooning is when a row of trucks communicate wirelessly, driving very closely behind one another in a convoy, ultimately without the need for a driver in each truck. The result being a far more fuel-efficient journey as there is less drag produced by each vehicle.

Another example would be Chatbots. „Chatbot“ is the new marketing buzzword that is steadily taking over from the previous fanfair around „Content Marketing“. It’s not that one has become more important than the other, but artificial intelligence (A.I) is leading to a new age of marketing – a conversational age – that brands and the businesses behind them need to fully understand and continually adapt to. Brands such as Starbucks and British Airways, are becoming evermore chatbot-savvy via Facebook’s Messenger or Slack. Smartphone Apps such as Duolingo are taking full advantage of the chatbot opportunity offering people new, more natural ways of learning a language through dialogue, albeit a text-based dialogue with a computer.

 

Machines are leading to more natural human interactions

ChatBots are actually just the tip of the iceberg. More generally, conversational interfaces, powered by machine learning, are rapidly increasing in popularity, as they allow people to achieve things without even lifting a finger (Think Apple’s Siri, Microsoft’s Cortana, or Amazon’s Alexa). Why pick up your phone, open your pizza delivery app and faff around with submitting your order, when you can just talk to your Amazon Echo and ask Alexa to do the same? No phone, no screens, just voice. In fact, our device screens are an incredibly impractical and unnatural way of interacting with the world. What could be more intuitive than interacting through conversation? There’s no need to learn a specific graphical interface or operating system. Just by talking naturally, your every wish becomes their command. As machine learning gets smarter and voice recognition becomes more accurate we could be freed from our small-screen-obsession, at least in the comfort of our own homes. The paradox is that A.I. is leading to our interactions becoming more human.

 

Man vs machine

The flip-side of it all is that as transportation, manufacturing, and marketing jobs are becoming more automated, we are in danger of making ourselves redundant. As has already been written by many others, if the trucks drive themselves, what do the drivers do instead? If the cars are built by robots, what do the engineers do instead? And if you’re thinking your job could never be done by a robot, then think again. Lawyers, journalists, accountants, investment bankers and doctors are also being threatened by the rise of automation. A “robot lawyer” chatbot has overturned 160,000 parking fines and is now helping refugees claim asylum. Robo-surgeons can already cut a more accurate incision better than the shaky hand of an ageing human surgeon. In 2012 the co-founder of Sun Microsystems predicted that 80% of medical doctors jobs would be lost within the next 20 years due to automation. Arguably, automation is in part fuelling the recent backlash to globalisation, due to more technological unemployment.

But it’s certainly not all bad. In fact, we need these kind of robotic developments to keep up with our exploding population worldwide. According to The Palmer Group, even if every person that wanted to become a doctor achieved their ambition, there still wouldn’t be enough doctors to serve the population. Robots and machines are adding great value here. In addition, according to an official European Commission study of the impact of robotic systems on employment in the EU (revised at the end of 2015), the use of industrial robots does not have significantly negative effects on employment and in fact produces superior efficiency in their manufacturing processes.

The robotic Armageddon depicted in Terminator 2 is thankfully not on the immediate horizon, but at a societal level, there are some serious questions that need to be answered about how far we want to go into an automated world and how we can regulate it enough, to ensure all jobs aren’t lost to machines, whilst still embracing the technological advancement, its value and efficiencies.

 

Manu factum vs. machina factum 

Ultimately, we, the consumers of goods, hold the power to say what we believe is ethical and what not; what we will buy and what we will avoid, but that requires brands to provide greater transparency around the production processes behind the goods we buy. In the same way we currently have the choice of buying goods from organic and ethical sources, in future we should be offered the same choice when it comes to human or machine-produced goods. It’s not totally outlandish to hope for a new label appearing on the goods we buy, not just organically-certified but also human-certified.

The term manufacturing, has come to mean „the making of articles on a large scale with machinery“ but perhaps we need to reclaim the Latin origins of this term: „manu factum“ – made by hand – and use it as a transparent & visible mark that let’s us know whether what we’re buying/interacting with is truly man-made, or at least in part. In future, this could and perhaps should also apply to customer services. Like something out of Blade Runner, we may be questioning if we are talking to a human or a replicant that is “more human than human”. It’s not that machine is bad and human is good, but we need the opportunity to make informed decisions.

As consumers, if we know for sure that the product or service is „manufactured“, in the original sense of the word, we will then be faced with the question: are we willing to pay extra for it?

 

It’s time for man to do some machine learning

Whilst this article expresses a number of concerns and potential dangers, ultimately technology has the power to augment human input rather than replace it. As brand and business shapers, we must do some deep machine learning of our own and continually re-evaluate the evolving opportunities and implications that go with evermore sophisticated artificial intelligence.

We need to adapt now to ensure our businesses are evolving in a sustainable way, embracing and supported by robotics. To quote the authors of “Race Against The Machine “the key to winning the race is not to compete against machines but to compete with machines”.

 

If you’re interested in identifying the latest opportunities and threats of automation or harnessing the conversational age for your brand and business, we have a team of humans who are enthusiastic about supporting your team around these topics. Just drop us a line.