Germany’s platform for artificial intelligence “Lernende Systeme”

It’s greatly inspiring and an honour being part of this unique platform which is concentrating knowledge and illustrating perspectives in the field of artificial intelligence and self learning systems

The Plattform Lernende Systeme brings together expertise from science, industry and society for fostering Germany‘s position as an international technology leader. It understands itself as a forum for exchange and cooperation.

Click here for more insight views:https://www.plattform-lernende-systeme.de/home-en.html

Designing self-learning systems for the benefit of society is the goal pursued by the Plattform Lernende Systeme which was launched by the Federal Ministry of Education and Research (BMBF) in 2017 at the suggestion of acatech. The members of the platform are organized into Working Groups and a Steering Committee which consolidate the current state of knowledge about self-learning systems and Artificial Intelligence. They point out developments in industry and society, analyse the skills which will be needed in the future and use real application scenarios to demonstrate the benefit of self-learning systems. A Managing Office at acatech coordinates the work of the platform.

Concept and aims of the platform

Self-learning systems are increasingly becoming a driving force behind digitalisation in business and society. They are based on Artificial Intelligence technologies and methods that are currently developing at a rapid pace in terms of performance. Self-learning systems are machines, robots and software systems that learn from data and use it to autonomously complete tasks that have been described in an abstract fashion – all without specific programming for each step.

Self-learning systems are becoming increasingly commonplace supporting people in their work and everyday lives. For example, they can be used to develop autonomous traffic systems, improve medical diagnostics and assist emergency services in disaster zones. They can help improve quality of life in many different respects, but are also fundamentally changing how humans and machines interact.

Self-learning systems have immense economic potential. As digitalisation takes hold, they are already helping companies in certain sectors to create entirely new business models based on data usage and are radically changing conventional value creation chains. This is opening up opportunities for new businesses, but can also represent a threat to established market leaders should they fail to react quickly enough.

Developing and introducing self-learning systems calls for special core skills, which need to be carefully nurtured to secure Germany’s pioneering role in this field. Using self-learning systems also raises numerous social, legal, ethical and security questions – with regard to data protection and liability, but also responsibility and transparency. To tackle these issues, we need to engage in broad-based dialogues as early as possible.

Plattform Lernende Systeme brings together leading experts in self-learning systems and Artificial Intelligence from science, industry, politics and civic organisations. In specialised focus groups, they discuss the opportunities, challenges and parameters for developing self-learning systems and using them responsibly. They derive scenarios, recommendations, design options and road maps from the results.

Handelsblatt AI Summit: Revolutionary technology & the impact on our society

AI thought leaders met on the 21st and 22nd at the #HBAISummit not only to discuss the latest developments in machine and deep learning programming but especially real use-cases, trends in the start-up scene, the role of Germany and Europe in the AI race and the impact of AI on our society. Read more here

It has been a great pleasure discussing and being inspired by a highly engaged auditorium during two sessions:

AI in Finance

Fireside Chat: Es geht immer Meer

Handelsblatt KI-Summit – Thoughtleaders in AI meet in Munich

It’s a great pleasure having the chance to meet international experts and supporting the Summit as Speaker on AI in Finance and with an evening fireside chat about leadership, transformation and the sea.

HANDELSBLATT KI-SUMMIT

DETAILS & LINK TO THE SUMMIT
Fireside Chat in der Future Lounge about Leadership in times of transformation

KI-INsights Businessguide

Read interviews and articles from experts in the KI-Businessguide published by the Handelsblatt for the Summit and free to download on this website

It was a pleasure supporting the publication with reflection on AI developments in the Financial Industry called: “Google, how are my stocks doing and what to do?” AI in the financial sector – the next big thing?
AI in Finance Session in the Handelsblatt KI-SUMMIT
AI in Finance Session in the Handelsblatt KI-SUMMIT

New ways – Artificial Intelligence in Cyber Security

Defense against cyber attacks through new technologies

Author: Jochen Werne – published by Der Bank-Blog – 15 February 2019

Cyber crime has become a serious threat to business, politics and private individuals since a long time. New technologies based on the use of artificial intelligence might offer more security.

The fight against cyber threats has become significantly more complex for global government organisations, businesses, and individuals in recent years. Technical protection of IT systems and infrastructures and thus data security in the narrower sense are no longer the only issues. Companies, for example, need to address the much broader concept of information security.

Solutions based on artificial intelligence could prove helpful in the fight against cybercrime. According to a study by the IBM Institute for Business Value, the spread of intelligent, AI-based security solutions will increase significantly in the coming years.

Technical protective measures have long since been based on machine learning, for example, to identify spam or phishing e-mails or to record trends and anomalies in large amounts of data – both in data traffic within the corporate network and in its external connections.

Jochen Werne
Jochen Werne

AI systems for the identification of cyber attacks

In future, for example, systems might also be able to identify hidden channels in the corporate network through which cyber criminals attempt to acquire data. AI’s greatest strength, pattern recognition, enables automated detection of a wide range of anomalies and security incidents. For this purpose, however, AI-based systems must also learn to distinguish between common IT failures and cyber attacks. In addition, self-learning algorithms need to take internal corporate processes into account to come up with precise results.

In the near future, according to a forecast by Christian Nern, former Head of Security Software DACH at IBM Germany and today Partner at the Consulting firm KPMG, AI-based security analysis systems will be able to detect and fend off attacks proactively. Then, according to the former IBM security software chief, the confrontation between cyber criminals and security officers could possibly take place directly between the AI systems they use.

Germany as a pioneer country

Germany, which considers itself a pioneer country in the fields of learning systems and artificial intelligence, has already launched a platform for artificial intelligence on this topic initiated by the Federal Ministry of Education and Research (BMBF): “Learning Systems”. The platform with its 200 members brings together leading experts from science, business and society and deals with technological, economic and social issues relating to the development and introduction of learning systems on an interdisciplinary and cross-sector basis.

One of the seven working groups deals in particular with IT security, privacy, law and ethics. The composition of the topics in this group shows the interwoven culture-specific discussions that will later lead to scenarios, recommendations, guidelines and roadmaps.

Intelligent combination of available modules

As often in cyber security topics, there is no patent solution for the numerous questions and challenges. A company-wide risk management system, which establishes appropriate technical and organisational measures and also takes into account findings from psychology and cultural studies, seems to be a sensible way forward.

The right balance between security awareness and security, individual freedom paired with increased personal responsibility as well as support through technology and organisational structure is probably the most promising approach in the current state of research and technology to effectively meet the challenges for information and IT security.

Tradition meets modernity: Why more and more banks are relying on artificial intelligence

Artificial intelligence is finding its way into the highly regulated world of banking. And not only GAFA Silicon Valley high-tech companies see it as the technology of the future, but also FinTechs and established banks. How it came to this, what possibilities and limits there are at the moment and why humans will remain irreplaceable not only when it comes to money – the commentary

by Jochen Werne, innovation and transformation expert
Munich private bank Bankhaus August Lenz

Original published in German in the IT-Finanzmagazin (31 July 2018). Translation by DeepL

After “FinTech”, “Blockchain” and “Crypto”, “AI” is the new buzzword in the banking world. Whether chatbots in the digital customer center or self-learning algorithms for highly complex investment strategies are being discussed – the omnipresence of the term suggests that the integration of artificial intelligence into one’s own business model seems to be virtually vital.

Artificial intelligence and big data are currently the strongest and most vibrant innovation trends in the financial sector …

… was also one of the guiding principles of Prof. Joachim Wuermeling, board member of the Deutsche Bundesbank, in his speech on “Artificial Intelligence” at the second annual FinTech and Digital Innovation Conference in February 2018 in Brussels.

The choice of the conference venue, which like rarely any other city combines both a belief in progress and a deeply rooted European tradition, can hardly be more symbolic of the forthcoming change. In fact, the topic is by no means new: the development towards an increased use of so-called non-human intelligence is based on approaches from the 1940s – with the invention of the first computers

Artificial intelligence: revolution as a reaction to mountains of data?

But what is now possible in times of exponential technologies is in fact nothing less than a revolution. The financial industry is sitting on a valuable mountain of data, the extent of which is currently difficult to estimate. The maturing AI systems would not only make the preparation and processing of this data easier, but also much more cost-effective, faster and more targeted. Data already collected could become the most valuable raw material and a resource due to the technological leaps in the field of AI, which, in combination with the enrichment of external, non-structured data, must be “usable” in a meaningful way.

The industry is asked to use private data in a sensitive way for the benefit of the customer, – a goal that should certainly apply to all AI-based approaches.

To find meaningful regulations for the handling and the effects of the use of AI on society, economy and thus on our life and the work of tomorrow is the task of politics. The fact that this topic is taken very seriously is evident not only in national initiatives such as the German Platform for Artificial Intelligence “Learning Systems”, but also in the European Artifical Intelligence shoulder-to-shoulder approach, which is being pushed forward by France and Germany.

“Digital hand holding” in the event of a financial crash is not enough

At present, it is still too early to say which operational areas of the financial world will sooner or later be supported – in part or even entirely – by the use of AI systems. However, the financial crises of the past have shown this time and again:

Trust is crucial when it comes to money. Trust in the markets, the banking system and the human contact as an intermediary in a complex issue”.

However, the banking industry knows very well from its own experience how easy it is to loose customer’s trust. An experience that Mark Zuckerberg and Facebook recently also had to make in connection with the Cambridge-Analytica scandal. As with every new technology and every new approach, the same applies to the topic of “intelligent” systems: a lot of trust, coupled with half-knowledge and a big dash of emotionality results in a popular trend cocktail, which, however, bears a certain risk of headaches on the following day.

Jochen Werne

Jochen Werne is the authorized signatory responsible for Marketing, Business Development, Product Management, Treasury and Payment Services at Bankhaus August Lenz & Co. After two years as navigator of the sailing training ship ‘Gorch Fock’, the international marketing and banking specialist completed his studies as client coverage analyst at Bankers Trust Alex. Brown International and in Global Investment Banking at Deutsche Bank AG, he has worked on numerous projects in other European and American countries. In 2001, he joined Accenture as a Customer Relationship Management Expert in the Financial Services Division before joining Bankhaus August Lenz & Co. AG in Munich, where he has since been responsible for various areas of the institute. As part of the Innovation Leadership Team of the Mediolanum Banking Group, a member of the expert council of Management Circle and the IBM Banking Innovation Council, Jochen Werne is a keynote speaker at numerous banking and innovation conferences.

Human. Digital. Culture. Our life in times of exponential technologies

16 January 2019 – Author: Jochen Werne

Almost every day, experts in the media try to create a historical analogy for us in order to explain the dynamics and speed with which changes are taking place today at all levels of our lives – from private consumption and our working world to international politics. Often analogies are drawn to different decades of the 20th century. The prominent British historian and Harvard professor Niall Ferguson contradicts these comparisons and sees an analogy rather in the effects that the invention of the printing press in the 15th century had on our lives and on our society. Only that today the changes due to exponential technologies and the Internet take place much faster.

For us as the HUMAN Factor, these comparisons are incredibly important. In times of uncertainty, they help us to better assess the changes and thus at least maintain a certain reassuring feeling of security and explainability. However, if we do not succeed in setting the right filters in times of social media and “information overload”, we run the risk that this feeling of understanding does not materialize and that we all too easily become victims of supposedly simple explanations and “fake news”. Ferguson uses a striking example to illustrate that this is not a new phenomenon and that serious technological changes have also brought major and often turbulent changes to society. In times of the invention of book printing, knowledge was spread more cheaply and a broad part of the population gained access to higher education. One of the first books to be printed in large numbers was the Bible. But also other writings, like “Malleus Maleficarum” or in English the “Hammer of Witches” became famous. The “Fake News” book served to justify the persecution of witches, appeared in 29 editions and has been second place on the book bestseller list for 200 years.

At the latest since the end of the 1990s, since the mass “democratization” of the Internet, our lives have been shaped by the exponential progress of modern technologies. The associated digitalization – the DIGITAL Factor – is not only a technical and economic challenge, but also a societal one. However, the enlightened man began, not to accept everything that a “Beautiful New World”, sometimes reminiscent of Aldous Huxley’s novel, promises. This is shown by citizen projects such as the so-called “Charter of Digital Fundamental Rights” of the European Union.

The word “exponential” automatically hides the logical conclusion that change will take place even faster in the future. These changes affect almost every industry and what is seen today as a billion-dollar future market can quickly become a basic business with significantly lower costs and thus significantly lower profit margins tomorrow. The camera chip of our smartphones costs today only about two to three Euros, a Spotify subscription, and thus the access to an incredible amount of music, only a few Euros a month.

The conclusion for companies in the 21st century is simple: Those who do not understand these exponential dynamics of technical development or do not take them sufficiently into account in their business model can quickly lose touch – not only with customers but also with potential business partners. But why is it so difficult for us to correctly assess the development potential of the technologies? The answer: People think linearly. This is why technologies are usually overestimated at the beginning of their development, but tend to be underestimated in the long run. This was first described in 1965 by the Intel engineer Gordon Moore – later known as Moore´s Law, one of the essential theoretical foundations of the “digital revolution”. In times of exponential technologies, our society risks a split between the group of people with an affinity for digital and digital natives and a group of people who have growing difficulties with the speed of change of our time. The latter have not learnt to keep pace with fast-moving digital innovations due to their low affinity, age or lack of points of contact in everyday life.

Throughout history, new technological possibilities have always come with threatening concepts that have been published and discussed on all media channels available during this period. Today it is: “total transparency”, “transparent consumer”, “constant availability” or even job loss due to ongoing automation and artificial intelligence. At the social and state level, attempts are being made to counteract such fears, to increase competitiveness and to involve the population in the process of change. Two of the many good examples referring to Germany are the strategy on artificial intelligence put in place by the Federal Government and the Platform for Learning Systems initiated by the Federal Ministry of Education and Research.

It is important never to forget, that every change – even if the trigger is a rapidly developing technology – requires a certain time horizon to be implemented and to create broad acceptance. Here the “CULTURE Factor” often comes into play. One example is cash. While the Scandinavian countries, above all Sweden, are about to digitalize their payment systems to a large extent, in Germany currently about 80 percent of all transactions are carried out with cash.

In every business model, global trends need to be identified, changes need to be driven, and local conditions need to be taken into account in order to be successful in this market. The same formula applies to societal change. Especially when it comes to creating an agenda for the use of new technologies for the benefit of our society.

————–

Photo from Niall Ferguson and Peter Robinson discuss networks and hierarchies throughout history in this episode of Uncommon Knowledge. Quote from the New York Times article “‘The Internet Is Broken’: @ev Is Trying to Salvage It” by David Streitfeld

Human. Digital. Culture. Unser Leben in Zeiten exponentieller Technologien

16. Januar 2019 – Autor: Jochen Werne

Fast täglich wird von Experten in den Medien versucht für uns eine geschichtliche Analogie herzustellen, um die Dynamik und Geschwindigkeit zu erklären, mit der sich heute Veränderungen auf allen Ebenen unseres Lebens – vom privaten Konsum, unserer Arbeitswelt bis hin zu internationaler Politik – vollziehen. Oftmals werden hierfür Vergleiche zu den 1930ern oder 70ern gezogen. Der bekannte britische Historiker und Harvard Professor Niall Ferguson widerspricht diesen Vergleichen und sieht eine Analogie vielmehr in den Effekten, die die Erfindung der Druckerpresse im 15. Jahrhundert auf unser Leben und auf unsere Gesellschaft hatte. Nur, dass sich heute die Veränderungen durch exponentielle Technologien und das Internet wesentlich schneller vollziehen. Für uns Menschen – also den HUMAN Factor – sind diese Vergleiche unglaublich wichtig. Sie helfen uns in Zeiten der Unsicherheit, die Veränderungen besser einschätzen zu können und somit zumindest ein gewisses beruhigendes Gefühl der Sicherheit und Erklärbarkeit zu erhalten. Wenn es uns jedoch nicht gelingt in Zeiten von Social Media und medialem „information overload“ die richtigen Filter zu setzen, laufen wir Gefahr, dass sich dieses Gefühl des Verständnisses nicht einstellt und wir allzu leicht Opfer vermeintlich einfacher Erklärungen und „Fake News“ werden. Dass dies kein neues Phänomen ist und gravierende technologische Veränderungen auch große und oftmals turbulente Veränderungen auf die Gesellschaft mit sich brachten, macht Ferguson an einem prägnanten Beispiel fest. In Zeiten der Erfindung des Buchdrucks kam es zu einer  kostengünstigeren Verbreitung von Wissen und somit zur Möglichkeit, dass breite Bevölkerungsschichten Zugang zu höherer Bildung erlangten. Eines der ersten in großer Auflage gedruckten Werke war die Bibel. Doch erlangten auch andere Schriften, wie „Malleus Maleficarum“ oder zu deutsch der Hexenhammer Berühmtheit. Das eindeutige „Fake News Werk“ diente zur Rechtfertigung der Hexenverfolgung, erschien in 29 Auflagen und belegte für immerhin 200 Jahre den zweiten Platz der Bücherbestsellerliste. 

Spätestens seit Ende der 90er Jahre, seit der massenhaften „Demokratisierung“ des Internets, ist unser aller Leben durch den exponentiellen Fortschritt moderner Technologien geprägt. Die damit einhergehende Digitalisierung – also der DIGITAL Factor – ist nicht nur eine technische und ökonomische Herausforderung, sondern vor allem auch eine gesellschaftliche. Dass der aufgeklärte Mensch jedoch beginnt, nicht alles einfach unreflektiert hinzunehmen, was eine an Aldous Huxley erinnernde „Schöne neue Welt“ zu versprechen scheint, zeigen Bürgerprojekte, wie die sogenannte „Charta der Digitalen Grundrechte“ der Europäischen Union. 

Bereits im Wort „exponentiell“ verbirgt sich automatisch die logische Schlussfolgerung, dass sich Veränderungen in Zukunft noch schneller vollziehen werden. Diese Veränderungen betreffen nahezu jede Branche und was heute ein milliardenschwerer Zukunftsmarkt ist,kann morgen schnell zu einem Basisgeschäft mit deutlich geringeren Kosten und somit auch deutlich geringeren Gewinnmargen werden. Der Kamera-Chip unserer Smartphones kostet heute nur noch rund zwei bis drei Euro, ein Spotify-Abo, und somit der Zugriff auf eine unfassbar große Menge an Musik, nur wenige Euro im Monat. 

Die Schlussfolgerung für Unternehmen im 21. Jahrhundert ist sichtlich einfach: Wer diese exponentiellen Dynamiken von technischer Entwicklung nicht versteht oder nicht ausreichend in seinem Geschäftsmodell berücksichtigt, kann schnell den Anschluss verlieren – Anschluss an Kunden aber auch an potenzielle Geschäftspartner. 

Doch warum fällt es uns so schwer, das Entwicklungspotenzial der Technologien richtig einzuschätzen? Die Antwort: Menschen denken linear. Deswegen werden Technologien zu Beginn der Entwicklung meist überschätzt, langfristig aber tendenziell unterschätzt. Dies wurde 1965 durch den Intel Ingenieur Gordon Moore erstmals beschrieben – später bekannt als Moore´s Law, einer der wesentlichen Theoriegrundlagen der „digitalen Revolution“. 

Unsere Gesellschaft lebt in Zeiten exponentieller Technologien natürlich auch mit der Gefahr einer Spaltung zwischen der Gruppe digital affiner Bevölkerungsschichten und Digital Natives sowie einer Gruppe von Menschen, die wachsende Schwierigkeiten mit der Veränderungsgeschwindigkeit unserer Zeit hat. Letztere haben aufgrund geringer Affinität, teilweise des Alters oder fehlender Berührungspunkte im Alltag nicht gelernt, mit den schnelllebigen digitalen Innovationen Schritt zu halten. Mit allen technischen Möglichkeiten geistern zudem Begrifflichkeiten durch die Medien, die vielen Sorgen bereiten und Ängste schüren: „Totale Transparenz“, „gläserner Konsument“, „ständige Verfügbarkeit“ oder gar Arbeitsplatzverlust aufgrund anhaltender Automatisierung und Artificial Intelligence. Auf gesellschaftlicher und staatlicher Ebene wird versucht solchen Ängsten entgegenzuwirken, Konkurrenzfähigkeit zu steigern und die eigene Bevölkerung in den  Veränderungsprozess miteinzubeziehen. Zwei der vielen guten Beispiele in Deutschland hierfür sind, die von der Bundesregierung verabschiedete Strategie zu Künstlicher Intelligenz oder die vom BMBF initiierte Plattform für Lernende Systeme.                  

Es gilt bei jeder Veränderung – sei der Auslöser auch eine sich schnell entwickelnde Technologie – nie zu vergessen, dass es einem zeitlichen Horizont bedarf um Neues zu implementieren und eine breite Akzeptanz zu schaffen. Hier kommt der „CULTURE Factor“ oftmals ins Spiel. Ein Beispiel ist Bargeld. Während die skandinavischen Länder, allen voran Schweden, davor stehen ihre Bezahlsysteme weitgehend zu digitalisieren, werden in Deutschland aktuell noch rund 80 Prozent aller Transaktionen mit Bargeld durchgeführt. Es gilt also in jedem Geschäftsmodell globale Trends zu erkennen, Veränderungen zu treiben, jedoch auch lokale Gegebenheiten zu berücksichtigen, um in diesem Markt erfolgreich zu sein. Dieselbe Formel gilt für unsere gesellschaftlichen Veränderungen und das Ziel neue Technologien zum Guten für unsere Gesellschaft einsetzen zu können.

„It’s not information overload. It’s filter failure.“

Author Jochen Werne – Original in German at Bank-Blog – Translation by DeepL

Banks have to become guides for investors

Financial blogs, online communities & Co.: The ubiquitous flood of information can be both a curse and a blessing for bank customers and investors. Today more than ever, banks and savings banks must be “guides” for their customers.

Anyone who wants to invest money today gets a flood of information about the net. Dedicated private investors worldwide can be almost as well informed as professionals, in case of interest and sufficient know-how and time. Never before in the history of mankind have so many people worldwide had so much information at their disposal. Whether prices in real time, the latest assessments by analysts or experts, key figures on a security in an industry comparison, the diversity of opinion in a community – all this is available around the clock.

How intensively and purposefully this offer is used is another question. It can be assumed that only a minority is so urgently concerned with their investment or can be so concerned at all. In addition, the situation of the decision-maker is adversely affected by two factors: the excessive amount of unfiltered information and the classic behavioral finance problem.

Coping with the abundance of data and big data

Alvin Toffler, who brought the term “information overload” to prominence in his bestseller “Future Shock” in 1970, described the phenomenon and its consequences as follows:

“Information overload occurs when the input quantity of a system exceeds its processing capacity. Decision makers have a fairly limited cognitive processing capacity. Therefore, if information overload occurs, it is likely that a decrease in decision quality will occur.”

Consequently, it can be concluded that the wealth of information that accumulates every day can hardly be processed by a classic investor alone, let alone placed in the right context. In addition, there is an almost unmanageable and constantly growing mass of financial products for private customers. In short, a large proportion of investment decisions are therefore not made analytically correct, but spontaneously subjectively and emotionally, as the studies by Nobel Prize winners Kahneman and Tversky show.

Banks as a guide only for wealthy clients?

Clay Shirky, writer and consultant for Social and Economic Effects of Internet technologies and well-known in New Media circles, presented an interesting reflection on the problem of information overload:

„It’s not information overload. It’s filter failure.“

This provides a great opportunity for traditional banks to position themselves as problem solvers for the investor. The alternative – at least for wealthy private clients – apart from filtering their own information mass, is dialogue with a competent expert. A person they trust – and trust to not only filter out from the wealth of data what is relevant to their needs, but also to protect them from the classic emotional mistakes of financial decisions in volatile markets.

Development of a new, adequate support concept for all bank customers

All those who do not belong to this clientele, and this is mostly the classic retail customer, have little more than to accept the zero interest rate on their accounts and short-term securities. According to Bundesbank statistics, the majority of Germans have invested their assets in these forms of investment. This makes the Federal Republic a leading nation in financial matters when it comes to missed opportunities, as can be read again and again in the Sunday editions of the major national daily newspapers.

Today more than ever, the function of bank and financial advisors must be to act as filters and guides for customers in the jungle, providing them with a flood of information. Because no algorithm, no digital advisory service can protect the investor from an ill-considered, intuitive and possibly wrong decision. For modern bank management, this means setting up a completely different support concept with cost-efficient consulting structures, a powerful and highly flexible team and the appropriate digital and mobile equipment.