applicationAccording to a study of the Ludwig-Maximilians-Universität Munich we are already at the point where readers can’t recognize anymore if an article is written by a human being or artificial intelligence (AI). On the contrary, the texts of robots had even greater credibility in the test group, the study revealed. But writing texts is not the only creative area in which AI advances. More and more start-ups feed their servers with data and build innovative services thanks to machine learning.

I see especially a real win in AI when tackling the biggest problem of digital media right now, namely fake news. Fact-checking tools use state-of-the-art software to review factual claims or linguistic subtleties. Besides that, the automatic verification of sources is as well conceivable with AI and the list of new tools could be continued at will: chatbots, speech recognition, translations and bug fixes are also among the areas of application, and AI is developing rapidly in many business areas.

The following gives a glimpse into AI’s potential with a few examples:

Personalized content in e-commerce: Today's marketers face the challenge to not only have to provide consumers with relevant content, but to play that content in the right place at the right time. For Episerver, AI-based algorithms analyze the entire content of the enterprise. The AI automatically tags the content with metadata and uses these tags, the integrated data analysis and the real-time analysis of the behavior of the website visitor, the content is then played. Users are addressed in a personalized way at various contact points on their digital customer journey that considers what visitors do on the website. Not to mention that content can also be optimized using AI. At Acrolinx, e.g. authors get supported in writing. The AI analyzes the grammar, tone of voice and terminology directly when entering the customer's specifications. The result is a content that is tailored to the brand and individuality of the customer.

Recommendations: Synchronize recommendations across different channels, such as websites, mobile web, native apps, or emails, and combine different types of referrals, such as displaying favorite, recently purchased, similar, complementary, or special promotions in any combination. All that can easily be done with the help of the Intelligent Content Engine, which can offers even much more, such as customer segmentation, automated personalization, real-time testing and optimization. Machine learning algorithms deliver personalized recommendations across all channels. With dynamically personalized content you convince your customers in real time while strengthening customer loyalty. The content is sent to each individual customer at the right time and channel (email, web, mobile shops), as e.g. Emarsys demonstrates.

Price optimization: We all know that the price affects the sales revenue effectively and strategically. If the assortment gets too big, it might happen to lose track, but software tools can help here. In order to find the optimal price, online shops like to use pricing tools. Prudsys offers the optimization of pricing processes for the entire product range at all touchpoints. Different optimization methods are used: (partially) automated pricing, inventory optimization, markdown optimization/pricing, single price optimization, etc.

Scan documents quickly: Leverton, founded in Berlin in 2012, has developed a self-learning algorithm that can scan hundreds of pages of a document in no time and which is used by the fund company Union Investment to manage its real estate portfolios. The software can read the contracts and analyze them with artificial intelligence, so that the portfolio manager no longer must read the complete contracts but sees the relevant paragraphs in the document.

Virtual consultants & FAQ Center: As clever guides for suitable products, the "smart" shopping bots on WhatsApp or Facebook Messenger can increase traffic and sales as well as customer satisfaction. Besides that, AI can help optimize inventory, which means out-of-stock rates can be reduced by up to 80% and revenue and profits increased by more than 5%.

Finding suitable music for brands: The team of the project ABC_DJ achieved a breakthrough in music research recently by providing a brand with a solution to accurately choose music that fits its personality and resonates with its audiences. At the push of a button, from tens of thousands of music titles, those that match a brand can easily be found – hence, audio branding in real time is now possible.

Providers using such intelligent systems have a clear advantage over competitors and can survive on the market, still it is all about people when it comes to decisive implementation: you need qualified employees and service providers with experience and trust.

By Daniela La Marca