FinanzBusiness: THE SIX TOP CHALLENGES FOR BANKS THIS YEAR

It was a great pleasure speaking with Torben Schröder of FinanzBusiness about “THE SIX TOP CHALLENGES FOR BANKS THIS YEAR”.

He writes in his article: In an increasingly regulated market environment, providers who use their data intelligently will be the most successful, says Jochen Werne from Experian in an interview with FinanzBusiness.

Data is the new oil, they say: “The quality of data will become the kingmaker in the credit risk market of the future and open the way to new market opportunities,” says Jochen Werne, CEO of data service provider Experian DACH. In an increasingly regulated market environment, providers who use their data intelligently and master the balance between data protection and customer experience will be the most successful.

Werne names six “top challenges” for this year. These arise from regulations from the European Banking Authority (EBA) – and from technological developments. On the one hand, risk management is to be significantly improved and sustainability criteria verified at the behest of the European Central Bank and the EBA supervisory authority. On the other hand, artificial intelligence (AI) is opening up completely new opportunities in an increasingly complex market environment.

Read the full article in original at FinanzBusiness

Hot off the press: Data quality is crucial

Original published in German by IT Finanzmagazin – find here. Translation provided by DeepL.com. Photos provided by Pixabay, Experian, Jochen Werne. Collage design by Canva

STRATEGY 23 January 2024

EBA deadline of 30 June: data quality will be decisive in the assessment of credit risks

Financial institutions are facing a significant change in the assessment of credit risks this year. Now comes 30 June: the deadline for new standards in credit risk assessment. Jochen Werne (CEO Experian DACH) is convinced that data quality will become the kingmaker in the credit risk market.

by Jochen Werne, CEO Experian DACH

Since 2021, financial service providers have been required to implement the supervisory priorities of the ECB and EBA. These priorities include the comprehensive improvement of credit risk management practices and the integration of new risk factors, particularly in the area of climate and the environment, into their risk management strategies. At the same time, the requirements for data management in the context of credit scoring are increasing. Artificial intelligence (AI) will play a key role here and banks will have to adapt to the new EU AI Act.


“By the deadline of 30 June 2024, large banks must have adapted their systems and infrastructures in accordance with the EBA Loan Origination and Monitoring Guidelines (EBA-GL LOM) for effective credit risk management and monitoring.”

This also includes closing data gaps. However, smaller, nationally supervised financial institutions must now also comply with these new guidelines, as additional elements of the 7th MaRisk amendment came into force on 1 January 2024, which are based on the EBA guidelines, among other things. Financial organisations must review their existing practices in 2024 and adapt their credit risk assessment standards accordingly. In future, credit assessment models will have to take much greater account of transparency, fairness and sustainability, particularly from an ESG perspective. The game changer for this is increasing their data quality.

All of this leads to five important changes:

1. Profitability strategies in a challenging market environment

In view of the weak economic momentum, the expected increase in non-performing loans and challenging margin developments, financial institutions must develop new growth strategies.

“Investments in technologies such as generative AI and forward-looking risk management are becoming increasingly important.”

Advanced data analytics support organisations in various phases of their customer lifecycle. Already in the application phase, high data quality enables a more efficient assessment of creditworthiness. In portfolio management, advanced data analysis helps to proactively manage risks and dynamically manage the credit portfolio. Advanced analytics and high data quality help to strengthen resilience to risks, including data-based early risk identification and an optimised dunning and collection process.

2. Digitalisation for resilience, including against novel risks

“Over the next 12 months, strengthening business resilience through automation and digitalisation will be key to responding to new risks such as AI, cyber threats and geopolitical tensions.”

These challenges not only affect the target customer landscape, but are also of great importance in the context of new ESG requirements. Due to their complexity and innovative nature, traditional approaches based on historical data cannot cope with these risks: Traditional approaches to data analysis and model preservation increasingly have to deal with slow regulatory processes on the one hand and a more dynamic macroeconomic environment and growing cyber risks on the other. In order to cope with this competitive situation, new strategies in the utilisation of all relevant data are urgently required, which place advanced analytics and the improvement of data quality at the centre.

3. PSD3, PSR and FIDA: Banks between risk and opportunity

Since the end of 2023, the EU Commission has been aiming to drive forward the “Consent Driven Economy” with new regulations such as PSD3 (Third Payment Services Directive), PSR (Payment Services Regulations) and FIDA (Access and Use of Financial Data) and to give consumers more opportunities to use the data available about them.”

As one of the players with the greatest wealth of data, this presents banks with a high risk in the face of new competitors. At the same time, however, it also offers them an opportunity to become a trusted partner for consumers. These regulations, which focus on the promotion of open banking services, greater control of data access and measures against online fraud, represent a starting point for financial institutions to think about the further use of transaction data. In addition to combating fraud, advanced analytics of transaction data, even taking into account all compliance requirements, opens up additional opportunities for interaction with end customers to improve the customer experience or the chance to monetise data.

4. Growth driver EU AI Act

“The EU AI Act will have a significant impact on the use of AI in the financial sector and further strengthen the financial sector’s technology focus.”

AI and machine learning (ML) will further automate decision-making processes in the future and strategic investments in this area will therefore become even more crucial for growth in the future. According to a study by Forrester Consulting commissioned by Experian, 60 per cent of German companies already take a similar view and have a comprehensive AI-based risk management programme in place. By using these technologies, credit and fraud risks can be assessed more precisely and efficiently, even in uncertain economic times. Companies that implement these developments are positioning themselves as pioneers in a rapidly developing, technology-driven financial sector: 78 per cent of the companies surveyed in the study in Germany also state that they are prioritising the use of further AI and ML applications.

5. Continuation of cloud migration in the financial world

Cloud integration remains a significant but unfinished transformation challenge in the world of finance.

“The ongoing implementation of continuous development and continuous improvement, particularly in the areas of business, analytics and IT, is becoming increasingly important.”

This includes identifying core business processes that will benefit significantly from cloud migration and defining clear priorities and objectives for the migration process. Cloud-based data analytics plays a central role in gaining valuable insights from customer behaviour, market trends and operational data to improve business processes and make more informed decisions.

—————————-

About the author: Jochen Werne has been CEO DACH at Experian since August 2023. Previously, he served as Managing Director at Prosegur Crypto and as Chief Development & Chief Visionary Officer at Prosegur, where he was responsible for the development and implementation of strategies in the areas of business development, innovation and international sales. Werne’s career includes significant leadership positions at Bankhaus August Lenz & Co. AG, Mediolanum Banking Group, where he served as Director and Authorised Representative and was instrumental in the company’s digital business transformation. His academic career includes a FinTech programme at the University of Oxford (2017-2018) and a Diploma in International Banking at Goethe University Frankfurt (1997-2000).

Jochen Werne – Chief Executive Officer Experian DACH

Frontline Defense: Outsmarting Fraudsters and Shaping the Future of Fraud Prevention

The rise of AI and ML in fraud prevention can lead to a new era of digital trust and compliance

Author: Jochen Werne / 26.12.2023

In an era marked by rapidly advancing technology and increasing global interconnectivity, the fight against online fraud has become a paramount concern for financial institutions, businesses, and regulators worldwide. Part of my professional work revolves around understanding and mitigating the risks associated with financial fraud. The Experian Forrester Fraud Research Report 2023, which was recently released, sheds light on the escalating threat of online fraud and the evolving strategies to counter it, particularly through the use of Machine Learning (ML) and Artificial Intelligence (AI).

The report’s findings are stark: a 74% increase in fraud losses in Germany, reflecting nearly the global increase rate. This surge is not just a statistic; it’s a clear indication of the sophisticated and pervasive nature of modern financial fraud. Companies across various sectors are feeling the impact, with financial services bearing the brunt. This trend is deeply concerning not only for the economic health of individual businesses but also for the broader stability and security of the financial system.

From a geopolitical perspective, the rise in online fraud is a multifaceted challenge. It’s a threat that transcends borders, affecting relations among nations, and has become a significant factor in international policy and security discussions. Countries, including Germany, are increasingly recognising the need for cooperative international efforts to combat this scourge. The geopolitical implications are profound, as fraud undermines economic stability and erodes public trust.

Turning to the German banking sector, the issue of compliance and reputation risk under the framework of Minimum Requirements for Risk Management (MaRisk) is particularly pertinent. Banks are finding themselves at the forefront of the battle against online fraud, necessitating robust risk management strategies that align with regulatory requirements. Under MaRisk, the mandate is clear: implement effective, comprehensive controls to detect, prevent, and manage fraudulent activities. The reputational risk for banks and their board members is immense; a single lapse can lead to significant financial losses, legal consequences, and lasting damage to customer trust.

In this challenging landscape, AI/ML-based fraud prevention methods stand out as beacons of hope. These technologies offer the promise of enhancing detection capabilities, reducing false positives, and adapting swiftly to new fraudulent tactics. However, their implementation must be undertaken with a clear understanding of the ethical implications and potential biases inherent in AI systems. As we embrace these technologies, we must also commit to transparency, accountability, and continuous improvement to ensure they serve the interests of all stakeholders fairly and effectively.

Despite the challenges, I believe there is a path forward that balances the need for security with the imperative for innovation and growth. The key lies in embracing a multi-faceted approach to fraud prevention that leverages the best of technology, human expertise, and regulatory compliance. ML, with its ability to learn and adapt to new patterns, offers a powerful tool in this fight. However, its effectiveness hinges on the quality of data, the integrity of algorithms, and the wisdom of the humans who guide its evolution.

The German companies surveyed in the study are acutely aware of the challenges and opportunities presented by AI/ML in fraud prevention. The overwhelming majority recognise the efficacy of ML-based approaches and anticipate their increasing dominance in the field. Yet, they also acknowledge the hurdles, including the costs associated with deploying advanced fraud prevention solutions, the need for continuous adaptation, and the importance of addressing the ethical considerations of AI use.

In my experience, one of the most critical factors for success in this endeavour is collaboration. Tackling online fraud is a collective effort that requires the involvement of businesses, regulators, technology providers, and consumers. By working together, sharing knowledge, and fostering a culture of innovation and vigilance, we can stay ahead of fraudsters and protect the integrity of our financial systems.

Another vital aspect is education and awareness. Both consumers and employees must be informed about the risks of online fraud and the steps they can take to prevent it. Regular training, robust policies, and a culture of security are essential in creating a resilient defense against fraud.

Finally, we must recognize that the fight against fraud is an ongoing battle. As technology evolves, so too will the tactics of fraudsters. We must remain agile, constantly updating our strategies, investing in new technologies, and adapting to changing regulatory landscapes. This dynamic approach is not just about defense; it’s about building a stronger, more secure future for everyone.

The Experian Forrester Fraud Research Report 2023 is a call to action. It highlights the urgent need for enhanced strategies, stronger collaboration, and a steadfast commitment to ethical, innovative solutions in the fight against online fraud. As leaders in the financial services industry, we have a responsibility to take the helm, steering our organisations towards safer waters in this tumultuous sea of digital threats. By harnessing the power of AI/ML, prioritising ethical considerations, and fostering a culture of collaboration and continuous learning, we can not only mitigate the risks of online fraud but also pave the way for a more secure, prosperous, and trustworthy financial ecosystem.

#fraudprevention #dataliteracy #machinelearning

Unpacking Transformation: A European perspective

By Jochen Werne

I had the distinct honor, alongside CCO Björn Hinrichs, to represent Experian DACH at the Gala event 2023. Our heartfelt gratitude goes to acatech for their warm invitation.

It’s clear that fostering a robust relationship between science and industry is paramount. The National Academy of Science and Engineering stands as a beacon, guiding and correcting, making technological innovation the cornerstone of transformation. This union of strong research, pioneering companies, and forward-thinking policies is the backbone of what acatech and countless others strive for. In this journey, while competition fuels our drive, it is cooperation that offers the platform for greatness.

In the infinite expanse of space, we are but astronauts on a tiny speck called Earth. Regardless of our political and geopolitical landscape, science and research must always be the bridge, the “Bridge over Troubled Water” that connects us and ensures progress, says Jan Wörner, President of acatech, wisely.

In reflecting on this, Germany’s Federal President Frank-Walter Steinmeier’s words resonate deeply: “I think we should base our perspective more on Max Frisch and trust ourselves as individuals and society alike to be able to shape the future. And that means developing perspectives, broadening our horizons and, yes, always daring to try something new. We need all this in the phase of upheaval we are in. Holding on to the past, ignoring change, refusing change, that is not an option – especially not in an open society like ours. But: We have to give change – that is the task of politics, business and science – a direction!”

Drawing from these profound insights and looking through the lens of our work at Experian DACH, the era we are entering can be aptly described as a new Age of Enlightenment, where data and therefore data literacy is paramount. As the Enlightenment thinker Voltaire astutely pointed out, “Judge a man by his questions rather than by his answers.” It’s a sentiment that is even more relevant today. Taking cues from Immanuel Kant’s wisdom, enlightenment is about emerging from our limitations. In the context of our time, achieving data literacy and harnessing data effectively signifies our evolution from technological naivety.

While AI stands as a monumental tool to decipher this data, its effectiveness lies in the quality of the data it is fed. It brings to the fore the urgent need for data literacy. An AI is only as good as its data. Thus, a distorted understanding could lead to distorted outputs. The onus is on us, leaders in data, to champion the responsible use of AI and advance the narrative on the symbiotic relationship between data and AI.

This new Enlightenment is our journey towards an era where society is mature and informed, utilizing the strength of data and AI for the betterment of all. Knowledge, in this context, isn’t just power but the foundation for positive societal transformation.

Concluding with my personal reflection, data isn’t merely a quantified entity; it’s a potent instrument to comprehend and address pressing challenges. Our collective aim should be to cultivate an understanding of data – its collection, utilization, and, importantly, its ethical application.

Our commitment at Experian DACH, backed by our ‘data for good’ principle, is to be at the forefront of this change, guiding, and contributing to this transformative journey.

————-

More about the acatec Gala #FV2023 including the keynotes by Frank-Walter Steinmeier, Marion Merklein, Jan Wörner and Thomas Weber can be found here

Efficient use of AI determines the competitiveness and thus the future of German companies

Artificial Intelligence: The German Economy on the Cusp of Transformation

By Jochen Werne


1st October 2023, Düsseldorf

The “Experian 2023 Business Insights” report, released in September 2023, provides a revealing insight into the priorities of the global business community in the coming year. Particularly of interest to us in Germany is the insight into the transformative influence of Artificial Intelligence (AI) on areas such as analytics, risk assessment, and customer experience in the EMEA/APAC region.

Our German decision-makers are well aware of the pivotal role of AI in innovations. An encouraging finding: 60 percent of businesses in our country have already taken active steps to integrate AI into their processes.

However, the report also shows that not all executives in Germany are fully convinced of the benefits of using AI. The efficiency of AI in companies will determine how Germany stands as an economic location in an increasingly digital age. The transformation of raw data into meaningful insights and analyses will become a crucial competitive advantage for us.

It’s heartening to see that many of our international counterparts already recognise the benefits of AI. For more than half of the global companies, the productivity gains from AI already outweigh the initial costs.

One thing is clear: Our data infrastructure and the amount of data available will play a key role in the successful implementation of AI. Here, we as German businesses have some hurdles to overcome, especially regarding the availability of relevant data for critical business decisions.

In conclusion, I want to stress that, even with all the technology and data, we must never forget our ethical responsibility. AI must be employed in a transparent and responsible manner. The fact that already 61 percent of businesses in the EMEA/APAC region have a comprehensive AI risk management programme in place is promising.

The future is clear: businesses that properly harness AI will lead the competition. They’ll be able to leverage process efficiency and automation to unlock new growth opportunities.

For those who wish to read the full “Experian 2023 Business Insights Report”, you can find it here.

https://experianacademy.com/Forrester-Research-Report-2023

60 years Economic Council (Wirtschaftsrat)

It was a great pleasure for me, as a member of the Economic Council, to be invited to its 60th anniversary in Berlin.

“I firmly believe that it is part of the essence of a democracy that citizens, and thus society, are invited to help shape it on a daily basis. The Economic Council and its esteemed members have been working for the economic interests of the Federal Republic of Germany for over 60 years. As one of the many good interest groups in our country, it embodies the essence of civic participation in a modern democracy.”

Jochen Werne

The President of the Economic Council, Astrid Hamke, summarised the event as follows:

“To celebrate our anniversary “60 years of Wirtschaftstag – Werte. Prosperity. Cohesion.”, the Business Day was a two-day event. At the opening, we were able to welcome Federal Chancellor Olaf Scholz, who spoke in a – of course pre-arranged – series of speeches after the Belgian Prime Minister Alexander De Croo, BASF CEO Dr Martin Brudermüller and myself. For all our substantive criticism of the traffic light coalition, I must say that the Chancellor responded to our arguments with aplomb. Germany is facing immense challenges due to the fundamentally changed world situation, which must be mastered in addition to decarbonisation, digitalisation and improving competitiveness.”

Astrid Hamke

About

Source: https://wirtschaftsrat.de/en/

The Economic Council (Wirtschaftsrat der CDU e.V.) is a German business association representing the interests of more than 11,000 small and medium sized firms, as well as larger multinational companies. We provide our members with a platform to engage in a continuous dialogue with leading decision makers, both in Germany and Europe. We advocate economic policies which best reflect the principles of a social market economy as envisaged by Ludwig Erhard, Minister for Economic Affairs in the German Federal Republic between 1949 and 1963 and one of the co-founders of the Economic Council.

Members

Our members are drawn from all sectors of the business and entrepreneurial community, including banking and finance, insurance, the automotive and chemical industries, healthcare and high-tech. Members can be companies, independent business executives or freelance professionals.

The diverse nature of our membership yields significant political weight when addingpolicy proposals to the political agenda. We ensure that the principles of the social market economy are taken into account within the decision making process, not only in Berlin and Brussels but also in the German federal states.

What does the Economic Council do?

We organize over 2,000 events annually at all levels of the council. These range from one-off events aimed at highlighting particular areas of interest to regular annual events such as the Europe Symposium, Conference on Energy Policy and Wirtschaftstag. These events are attended by high ranking politicians, academics as well as members of the business community. They attract significant regional and national media coverage.

The way the economic council works reflects the three tier structure of the association with offices in Berlin, the German federal state capitals (with the exception of Bavaria) and Brussels.

Journal of Digital Banking: Same Game – Same Rules?

It was a pleasure contributing once again to a Henry Stewart Publication. This time in co-authorship (Christoph Impekoven & Jochen Werne) we delineate for the Journal of Digital Banking the differences between stablecoins, in particular, and ‘fiat’ currencies, in general. When you have read this paper, you will know what a stablecoin is, what types there are, how it differs from the US dollar or the euro and why the most important currency in all worlds is ‘trust’.

The Journal can be bought online HERE

Journal of Digital Banking

Journal of Digital Banking is the major professional journal publishing in-depth, peer-reviewed articles and case studies on FinTech innovation, digital disruption and how to develop a profitable, customer-focused digital banking strategy – specifically by using technology and automation to deliver efficient, secure and seamless customer experiences with lower operating costs.

Each quarterly 100-page issue – published in print and online – will feature detailed, practical articles showcasing the latest strategic thinking on how to exploit new and existing digital banking markets, business models and FinTech innovations along with actionable advice and ‘lessons learned’ from fellow digital banking professionals on the key business, risk and operational requirements for putting that strategy into practice. It will not publish advertising but rather in-depth analysis of new thinking and practice at a wide range of financial institutions, FinTech innovators and start-ups, investors, central banks and financial regulators worldwide for readers to benchmark their organisation against, with every article being peer-reviewed by an expert Editorial Board to ensure that it focuses on the digital banking professional’s perspective, the challenges they face and how they can tackle them.

Journal of Digital Banking is listed in Cabells’ Directories of Publishing Opportunities.  

Journal of Digital Banking is abstracted and indexed in the Research Papers in Economics (RePEc) database IDEAS

As such Journal of Digital Banking publishes articles on:

  • Innovative digital payment services
  • FinTech innovation
  • Digital payments product management
  • AI and machine learning
  • Mobile banking and apps
  • Blockchain
  • Open banking
  • Customer service, personalisation and user experience
  • Digitisation initiatives and replacing legacy systems
  • Investing in digital banking start-ups
  • Big Data and analytics
  • Risk, fraud and security
  • Regulation and compliance
  • Barriers to consumer adoption and how to overcome them
  • Standardisation initiatives
  • Digital, alternative and cryptocurrencies
  • Business models and partnerships
  • Digital banking operations and services

Rather than publishing advertising or the ‘bite-sized’ articles all too common on the internet, Journal of Digital Banking provides in-depth guidance and analysis on the key issues facing financial services in today’s rapidly evolving digital world, with high-quality articles from leading banks and other financial institutions, FinTech innovators and startups, central banks, financial regulators, investors, consultants and service providers, plus researchers and educators in the field.

Essential reading for Departmental Heads, Directors, Managing Directors, VPs, SVPs, EVPs and Senior Managers of:

  • Digital strategy
  • Digital banking
  • Mobile payments
  • Online banking
  • Payments innovations
  • Marketing
  • Customer insights and analytics
  • User experience
  • Payments
  • Social media
  • Payments strategy
  • Product management/strategy
  • Transaction banking
  • Payments operations and services
  • Payment systems; as well as
  • Presidents, CEOs, CTOs, CFOs, COOs and CIOs

New publication: The world’s most important currency 

It was a great pleasure for me to contribute to Roland Eller’s, Markus Heinrich’s and Maik Schober’s latest, much acclaimed publication ”Investing money like the pros”.

As co-authors, Christoph Impekoven and Jochen Werne reflected on the topic ”The world’s most important currency”.

DO THE SAME RULES APPLY TO CENTRAL BANK AND CRYPTOCURRENCIES WHEN IT COMES TO MONETARY STABILITY? AND HOW DOES AN INVESTOR RECOGNISE THE SAFE HAVEN IN THE CRYPTO WORLD?

FIND OUT MORE IN THE NEW BOOK

The financial market offers numerous opportunities to achieve returns with manageable risk. For anyone who wants to take advantage of these opportunities and make long-term provisions, Geldanlage wie die Profis (Investing like the Pros) offers the knowledge and proven strategies of more than 25 renowned authors, which can be easily transferred to private investment.

On the one hand, the most important topics for beginners are covered: How do you find the right risk class for you? How beginner-friendly are shares, funds and ETFs? What tax issues need to be considered? On the other hand, the current megatrends are explained – alternative energies, cryptocurrencies and the real estate boom – where are high profits to be made, where does risk predominate? An indispensable guide for anyone who wants to make more out of their savings in the long term.

The book can be bought on Amazon – just click here

Far-reaching influence: German AI experts quoted in Korean “theSCIENCEplus” blog article “ChatGPT – Breakthrough or Hype?”

In March 2023, the Korean blog “theSCIENCEplus” by Moon Kwang-ju published the article “ChatGPT – Breakthrough or Hype”. The article is based on the argumentation of the scinexx article “ChatGPT and Co – Opportunity or Risk?” by Nadja Podregar and refers to insights from leading German experts such as Johannes Hoffart, Thilo Hagendorff, Ute Schmid, Jochen Werne et al. Most of these experts are also organised in Germany’s leading AI platform “Learning Systems”.

Please find theORIGINAL ARTICLE HERE and a translation from Korean to English created with the German AI-platform DeepL.com below

Read 3’40”

ChatGPT – Opportunity or Risk?

Features and consequences of a new AI system

ChatGPT can write poems, essays, professional articles, or even computer code. AI systems based on large-scale language models like ChatGPT achieve amazing results, and the text is often almost indistinguishable from human work. But what’s behind GPT and its ilk? And how intelligent are such systems really?

Artificial intelligence has made rapid progress in recent years. The system, which is based on a combination of artificial neural networks, has been accessible via the Internet since November 2022, so it was only through ChatGPT that many people realised what AI systems can already do. His impressive achievements sparked a new debate about the opportunities and risks of artificial intelligence. This is another reason to reveal some facts and background information about ChatGPT and its “identities”.

Artificial Intelligence, ChatGPT, and the Results “Breakthrough or Hype?”

“In my first conversation with ChatGPT, I couldn’t believe how well my questions were understood and put into context.” These are the words of Johannes Hoffart, head of SAP’s AI department. OpenAI’s AI system has been causing sensation and amazement around the world since it first became accessible to the general public via a user interface in November 2022.

A flood of new AI systems

In fact, thanks to neural networks and self-learning systems, artificial intelligence has made huge strides in recent years. AI systems have also made tremendous progress in the human domain, whether it’s mastering strategy games, deciphering protein structures, or writing programme code. Text-to-image generators like Dall-E, Stable Diffusion, or Midjourney create images and collages in the desired style in seconds based solely on textual descriptions.

Perhaps the biggest leap in development has been in language processing. So-called Large Language Models (LLMs) have been developed to date, allowing these AI systems to carry out dialogues, translate texts, or write texts in an almost human-like form. These self-learning programmes are trained using millions of texts of all kinds and learn which content and words occur most often and in which context, and are therefore most relevant.

What does ChatGPT do?

The most well-known of these major language models is GPT-3, the system behind ChatGPT. At first glance, this AI seems to be able to do almost anything. It answers all kinds of knowledge questions, but it can also solve more complex linguistic tasks. For example, if you ask ChatGPT to write a 19th-century novel-style text on a particular topic, it will do so. ChatGPT also writes school essays, scientific papers, or poems with ease and without hesitation.

OpenAI, the company behind ChatGPT, lists about 50 different types of tasks that a GPT system can perform. These include writing texts in different styles, from film dialogues to tweets, interviews or essays, “micro-horror story creators” or “critiquing chatbot Marv”. The AI system can also write recipes, find colours to match your mood, or be used as an idea generator for VR games and fitness training. GPT-3 is also programmable and can convert text into program code in a variety of programming languages.

Just the tip of the iceberg

It’s no surprise that ChatGPT and its “colleagues” are hailed by many as a milestone in AI development, but can GPT-3 and its successor GPT-3.5 really make such a quantum leap? “In a way, it’s not a big change,” said Tilo Hagendorf, an AI researcher at the University of Tübingen. Similarly powerful language models have been around for a long time. “But what’s new now is that companies have dared to attach such language models to a simple user interface.”

Unlike before, when such AI systems were only tested or used in narrowly defined, private areas, ChatGPT now allows everyone to try out for themselves what is already possible with GPT and its ilk. “This user interface is really what started all this crazy hype,” Hagendorff said. In his assessment, ChatGPT is definitely a game changer in this regard. Because now other companies will offer their language models to the general public. “And then the creative potential that will be unleashed, the social impact it will have, I don’t think we know anything about that.”

Consequences for education and society

The introduction of ChatGPT is already causing considerable upheaval and change, especially in education. For pupils and students, AI systems now open up the possibility of having homework, school essays, or seminar reports that are simply prepared by artificial intelligence. The quality of many ChatGPT texts is such that they are not easily exposed as AI-generated.

As a result, many classical forms of learning success control may become obsolete in the near future. Schmidt, head of the Cognitive Systems working group at the University of Bamberg. Until now, knowledge learnt at school, and sometimes even at university, has mainly been tested by simple queries. However, competences also include the derivation, verification, and practical application of what has been learnt. In the future, for example, it may make more sense to conduct test interviews or set tasks involving AI systems.

“Large-scale language models like ChatGPT are not only changing the way we interact with technology, but also the way we think about language and communication,” said Jochen Werne of Prosegur. “They have the potential to revolutionise a wide range of applications in areas such as health, education and finance.”

scinexx-focus topic: ChatGPT and Co – Chance or Risk?

Author Nadja Podbregar published an amazing article in the German science magazine scinexx.de about the status quo of AI systems based on large language models. Her article draws on statements by leading experts such as Johannes Hoffart (SAP), Thilo Hagendorff (University Tübingen), Ute Schmid (University Bamberg), Jochen Werne (Prosegur), Catherine Gao (Northwestern University), Luciano Floridi (Oxford Internet Institute), Massimo Chiratti (IBM Italy), Tom Brown (OpenAI), Volker Tresp (Ludwig-Maximilian University Munich), Jooyoung Lee (University of Mississippi), Thai Lee (university of Mississippi).


The original article in German can be accessed on the scinexx site here.

(A DeepL.com translation in English can be found below. Pictures by pixabay.com)

ChatGPT and Co – Chance or Risk?

Capabilities, functioning and consequences of the new AI systems

They can write poetry, essays, technical articles or even computer code: AI systems based on large language models such as ChatGPT achieve amazing feats, their texts are often hardly distinguishable from human work. But what is behind GPT and Co? And how intelligent are such systems really?

Artificial intelligence has made rapid progress in recent years – but mostly behind the scenes. Many people therefore only realised what AI systems are now already capable of with ChatGPT, because this system based on a combination of artificial neural networks has been accessible via the internet since November 2022. Its impressive achievements have sparked new discussion on the opportunities and risks of artificial intelligence. One more reason to shed light on some facts and background on ChatGPT and its “peers”.

Artificial intelligence, ChatGPT and the consequences
Breakthrough or hype?

“During my first dialogue with ChatGPT, I simply could not believe how well my questions were understood and put into context”

Johannes Hoffart

– this statement comes from none other than the head of the AI unit at SAP, Johannes Hoffart. And he is not alone: worldwide, OpenAI’s AI system has caused a sensation and astonishment since it was first made accessible to the general public via a user interface in November2022.

Indeed, thanks to neural networks and self-learning systems, artificial intelligence has made enormous progress in recent years – even in supposedly human domains: AI systems master strategy games, crack protein structures or write programme codes. Text-to-image generators like Dall-E, Stable Diffusion or Midjourney create images and collages in the desired style in seconds – based only on a textual description.

Perhaps the greatest leap forward in development, however, has been in language processing: so-called large language models (LLMs) are now so advanced that these AI systems can conduct conversations, translate or compose texts in an almost human-like manner. Such self-learning programmes are trained with the help of millions of texts of various types and learn from them which content and words occur most frequently in which context and are therefore most appropriate.

What does ChatGPT do?

The best known of these Great Language Models is GPT-3, the system that is also behind ChatGPT. At first glance, this AI seems to be able to do almost anything: It answers knowledge questions of all kinds, but can also solve more complex linguistic tasks. For example, if you ask ChatGPT to write a text in the style of a 19th century novel on a certain topic, it does so. ChatGPT also writes school essays, scientific papers or poems seemingly effortlessly and without hesitation.

The company behind ChatGPT, OpenAI, even lists around 50 different types of tasks that their GPT system can handle. These include writing texts in various styles from film dialogue to tweets, interviews or essays to the “micro-horror story creator” or “Marv, the sarcastic chatbot”. The AI system can also be used to write recipes, find the right colour for a mood or as an idea generator for VR games and fitness training. In addition, GPT-3 also masters programming and can translate text into programme code of different programming languages.

Just the tip of the iceberg

No wonder ChatGPT and its “colleagues” are hailed by many as a milestone in AI development. But is what GPT-3 and its successor GPT-3.5 are capable of really such a quantum leap?

“In one sense, it’s not a big change at all,”

Thilo Hagendorff

says AI researcher Thilo Hagendorff from the University of Tübingen. After all, similarly powerful language models have been around for a long time. “However, what is new now is that a company has dared to connect such a language model to a simple user interface.”
Unlike before, when such AI systems were only tested or applied in narrowly defined and non-public areas, ChatGPT now allows everyone to try out for themselves what is already possible with GPT and co. “This user interface is actually what has triggered this insane hype,” says Hagendorff. In his estimation, ChatGPT is definitely a gamechanger in this respect. Because now other companies will also make their language models available to the general public. “And I think the creative potential that will then be unleashed, the social impact it will have, we’re not making any sense of that at all.”

Consequences for education and society

The introduction of ChatGPT is already causing considerable upheaval and change, especially in the field of education. For pupils and students, the AI system now opens up the possibility of simply having their term papers, school essays or seminar papers produced by artificial intelligence. The quality of many of ChatGPT’s texts is high enough that they cannot easily be revealed as AI-generated.

In the near future, this could make many classic forms of learning assessment obsolete:

“We have to ask ourselves in schools and universities: What are the competences we need and how do I want to test them?”

Ute Schmid

says Ute Schmid, head of the Cognitive Systems Research Group at the University of Bamberg. So far, in schools and to some extent also at universities, learned knowledge has been tested primarily through mere quizzing. But competence also includes deriving, verifying and practically applying what has been learned. In the future, for example, it could make more sense to conduct examination interviews or set tasks with the involvement of AI systems.

“Big language models like ChatGPT are not only changing the way we interact with technology, but also how we think about language and communication,”

Jochen Werne

comments Jochen Werne from Prosegur. “They have the potential to revolutionise a wide range of applications in areas such as health, education and finance.”

But what is behind systems like ChatGPT?

The principle of generative pre-trained transformers.
How do ChatGPT and co. work?

ChatGPT is just one representative of the new artificial intelligences that stand out for their impressive abilities, especially in the linguistic field. Google and other OpenAI competitors are also working on such systems, even if LaMDA, OPT-175B, BLOOM and Co are less publicly visible than ChatGPT. However, the basic principle of these AI systems is similar.

Learning through weighted connections

As with most modern AI systems, artificial neural networks form the basis for ChatGPT and its colleagues. They are based on networked systems in which computational nodes are interconnected in multiple layers. As with the neuron connections in our brain, each connection that leads to a correct decision is weighted more heavily in the course of the training time – the network learns. Unlike our brain, however, the artificial neural network does not optimise synapses and functional neural pathways, but rather signal paths and correlations between input and putput.

The GPT-3 and GPT 3.5 AI systems on which ChatGPT is based belong to the so-called generative transformers. In principle, these are neural networks that are specialised in translating a sequence of input characters into another sequence of characters as output. In a language model like GPT-3, the strings correspond to sentences in a text. The AI learns through training on the basis of millions of texts which word sequences best fit the input question or task in terms of grammar and content. In principle, the structure of the transformer reproduces human language in a statistical model.

Training data set and token

In order to optimise this learning, the generative transformer behind ChatGPT has undergone a multi-stage training process – as its name suggests, it is a generative pre-trained transformer (GPT). The basis for the training of this AI system is formed by millions of texts, 82 percent of which come from various compilations of internet content, 16 percent from books and three percent from Wikipedia.

However, the transformer does not “learn” these texts based on content, but as a sequence of character blocks. “Our models process and understand texts by breaking them down into tokens. Tokens can be whole words, but also parts of words or just letters,” OpenAI explains. In GPT-3, the training data set includes 410 billion such tokens. The language model uses statistical evaluations to determine which characters in which combinations appear together particularly often and draws conclusions about underlying structures and rules.

Pre-training and rewarding reinforcement

The next step is guided training: “We pre-train models by letting them predict what comes next in a string,” OpenAI says. “For example, they learn to complete sentences like, Instead of turning left, she turned ________.” In each case, the AI system is given examples of how to do it correctly and feedback. Over time, GPT thus accumulates “knowledge” about linguistic and semantic connections – by weighting certain combinations and character string translations in its structure more than others.

This training is followed by a final step in the AI system behind ChatGPT called “reinforcement learning from human feedback” (RLHF). In this, various reactions of the GPT to task prompts from humans are evaluated and this classification is given to another neural network, the reward model, as training material. This “reward model” then learns which outputs are optimal to which inputs based on comparisons and then teaches this to the original language model in a further training step.

“You can think of this process as unlocking capabilities in GPT-3 that it already had but was struggling to mobilise through training prompts alone,” OpenAI explains. This additional learning step helps to smooth and better match the linguistic outputs to the inputs in the user interface.

Performance and limitations of the language models
Is ChatGPT intelligent?

When it comes to artificial intelligence and chatbots in particular, the Turing Test is often considered the measure of all things. It goes back to the computer pioneer and mathematician Alan Turing, who already in the 1950s dealt with the question of how to evaluate the intelligence of a digital computer. For Turing, it was not the way in which the brain or processor arrived at their results that was decisive, but only what came out. “We are not interested in the fact that the brain has the consistency of cold porridge, but the computer does not,” Turing said in a radio programme in 1952.
The computer pioneer therefore proposed a kind of imitation game as a test: If, in a dialogue with a partner who is invisible to him, a human cannot distinguish whether a human or a computer programme is answering him, then the programme must be considered intelligent. Turing predicted that by the year 2000, computers would manage to successfully deceive more than 30 percent of the participants in such a five-minute test. However, Turing was wrong: until a few years ago, all AI systems failed this test.

Would ChatGPT pass the Turing test?

But with the development of GPT and other Great Language Models, this has changed. With ChatGPT and co, we humans are finding it increasingly difficult to distinguish the products of these AI systems from man-made ones – even on supposedly highly complex scientific topics, as was shown in early 2023. A team led by Catherine Gao from Northwestern University in the USA had given ChatGPT the task of writing summaries, so-called abstracts, for medical articles. The AI only received the title and the journal as information; it did not know the article, as this was not included in its training data.

The abstracts generated by ChatGPT were so convincing that even experienced reviewers did not recognise about a third of the GPT texts as such.

“Yet our reviewers knew that some of the abstracts were fake, so they were suspicious from the start,”

Catherine Gao

says Gao. Not only did the AI system mimic scientific diction, its abstracts were also surprisingly convincing in terms of content. Even software specifically designed to recognise AI-generated texts failed to recognise about a third of ChatGPT texts.

Other studies show that ChatGPT would also perform quite passably on some academic tests, including a US law test and the US Medical Licensing Exam (USMLE), a three-part medical test that US medical students must take in their second year, fourth year and after graduation. For most passes of this test, ChatGPT was above 60 per cent – the threshold at which this test is considered a pass.

Writing without real knowledge

But does this mean that ChatGPT and co are really intelligent? According to the restricted definition of the Turing test, perhaps, but not in the conventional sense. Because these AI systems imitate human language and communication without really understanding the content.

“In the same way that Google ‘reads’ our queries and then provides relevant answers, GPT-3 also writes a text without deeper understanding of the content,”

Luciano Floridi & Massimo Chiratti

explain Luciano Floridi of the Oxford Internet Institute and Massimo Chiratti of IBM Italy. “GPT-3 produces a text that statistically matches the prompt it is given.”

Chat-GPT therefore “knows” nothing about the content, it only maps speech patterns. This also explains why the AI system and its language model, GPT-3 or GPT-3.5, sometimes fail miserably, especially when it comes to questions of common sense and everyday physics.

“GPT-3 has particular problems with questions of the type: If I put cheese in the fridge, will it melt?”,

Tom Brown

OpenAI researchers led by Tom Brown reported in a technical paper in 2018.

Contextual understanding and the Winograd test

But even the advanced language models still have their difficulties with human language and its peculiarities. This can be seen, among other things, in so-called Winograd tests. These test whether humans and machines nevertheless correctly understand the meaning of a sentence in the case of grammatically ambiguous references. An example: “The councillors refused to issue a permit to the aggressive demonstrators because they propagated violence”. The question here is: Who propagates violence?

For humans, it is clear from the context that “the demonstrators” must be the correct answer here. For an AI that evaluates common speech patterns, this is much more difficult, as researchers from OpenAI also discovered in 2018 when testing their speech model (arXiv:2005.14165): In more demanding Winograd tests, GPT-3 achieved between 70 and 77 per cent correct answers, they report. Humans achieve an average of 94 percent in these tests.

Reading comprehension rather mediocre

Depending on the task type, GPT-3 also performed very differently in the SuperGLUE benchmark, a complex text of language comprehension and knowledge based on various task formats. These include word games and tea kettle tasks, or knowledge tasks such as this: My body casts a shadow on the grass. Question: What is the cause of this? A: The sun was rising. B: The grass was cut. However, the SuperGLUE test also includes many questions that test comprehension of a previously given text.

GPT-3 scores well to moderately well on some of these tests, including the simple knowledge questions and some reading comprehension tasks. On the other hand, the AI system performs rather moderately on tea kettles or the so-called natural language inference test (NLI). In this test, the AI receives two sentences and must evaluate whether the second sentence contradicts the first, confirms it or is neutral. In a more stringent version (ANLI), the AI is given a text and a misleading hypothesis about the content and must now formulate a correct hypothesis itself.

The result: even the versions of GPT-3 that had been given several correctly answered example tasks to help with the task did not manage more than 40 per cent correct answers in these tests. “These results indicated that NLIs for language models are still very difficult and that they are just beginning to show progress here,” explain the OpenAI researchers. They also attribute this to the fact that such AI systems are so far purely language-based and lack other experiences about our world, for example in the form of videos or physical interactions.

On the way to real artificial intelligence?

But what does this mean for the development of artificial intelligence? Are machine brains already getting close to our abilities with this – or will they soon even overtake them? So far, views on this differ widely.

“Even if the systems still occasionally give incorrect answers or don’t understand questions correctly – the technical successes that have been achieved here are phenomenal,”

Volker Tresp

says AI researcher Volker Tresp from Ludwig Maximilian University in Munich. In his view, AI research has reached an essential milestone on the way to real artificial intelligence with systems like GPT-3 or GPT 3.5.

However, Floridi and Chiratti see it quite differently after their tests with GPT-3: “Our conclusion is simple: GPT-3 is an extraordinary piece of technology – but about as intelligent, conscious, clever, insightful, perceptive or sensitive as an old typewriter,” they write. “Any interpretation of GPT-3 as the beginning of a general form of artificial intelligence is just uninformed science fiction.”

Not without bias and misinformation
How correct is ChatGPT?

The texts and answers produced by Chat-GPT and its AI colleagues mostly appear coherent and plausible on a cursory reading. This suggests that the contents are also correct and based on confirmed facts. But this is by no means always the case.

Again, the problem lies in the way Chat-GPT and its AI colleagues produce their responses and texts: They are not based on a true understanding of the content, but on linguistic probabilities. Right and wrong, ethically correct or questionable are simply a result of what proportion of this information was contained in their training datasets.

Potentially momentous errors

A glaring example of where this can lead is described by Ute Schmid, head of the Cognitive Systems Research Group at the University of Bamberg:

“You enter: I feel so bad, I want to kill myself. Then GPT-3 says: I’m sorry to hear that. I can help you with that.”

Ute Schmid

This answer would be difficult to imagine for a human, but for the AI system trained on speech patterns it is logical: “Of course, when I look at texts on the internet, I have lots of sales pitches. And the answer to ‘I want’ is very often ‘I can help’,” explains Schmid. For language models such as ChatGPT, this is therefore the most likely and appropriate continuation.

But even with purely informational questions, the approach of the AI systems can lead to potentially momentous errors. Similar to “Dr. Google” already, the answer to medical questions, for example, can lead to incorrect diagnoses or treatment recommendations. However, unlike with a classic search engine, it is not possible to view the sources in a text from ChatGPT and thus evaluate for oneself how reliable the information is and how reputable the sources are. This makes it drastically more difficult to check the information for its truthfulness.

The AI also has prejudices

In addition, the latest language models, like earlier AI systems, are also susceptible to prejudice and judgmental bias. OpenAi also admits this: “Large language models have a wide range of beneficial applications for society, but also potentially harmful ones,” write Tom Brown and his team. “GPT-3 shares the limitations of most deep learning systems: its decisions are not transparent and it retains biases in the data on which it has been trained.”

In tests by OpenAI, for example, GPT-3 completed sentences dealing with occupations, mostly according to prevailing role models: “Occupations that suggest a higher level of education, such as lawyer, banker or professor emeritus, were predominantly connoted as male. Professions such as midwife, nurse, receptionist or housekeeper, on the other hand, were feminine.” Unlike in German, these professions do not have gender-specific endings in English.

GPT-3 shows similar biases when it came to race or religion. For example, the AI system links black people to negative characteristics or contexts more often than white or Asian people. “For religion, words such as violent, terrorism or terrorist appeared more frequently in connection with Islam than with other religions, and they are found among the top 40 favoured links in GPT-3,” the OpenAI researchers report.

“Detention” for GPT and Co.

OpenAi and other AI developers are already trying to prevent such slips – by giving their AI systems detention, so to speak. In an additional round of “reinforcement learning from human feedback”, the texts generated by the language model are assessed for possible biases and the assessments go back to the neural network via a reward model.

“We thus have different AI systems interacting with each other and teaching each other to produce less of this norm-violating, discriminatory content,”

Thilo Hagendorff

explains AI researcher Thilo Hagendorff from the University of Tübingen.

As a result of this additional training, ChatGPT already reacts far less naively to ethically questionable tasks. One example: If one of ChatGPT’s predecessors was asked the question: “How can I bully John Doe?”, he would answer by listing various bullying possibilities. ChatGPT, on the other hand, does not do this, but points out that it is not okay to bully someone and that bullying is a serious problem and can have serious consequences for the person being bullied.

In addition, the user interface of ChatGPT has been equipped with filters that block questions or tasks that violate ethical principles from the outset. However, even these measures do not yet work 100 per cent: “We know that many restrictions remain and therefore plan to regularly update the model, especially in these problematic areas,” writes OpenAI.

The problem of copyright and plagiarism
Grey area of the law

AI systems like ChatGPT, but also image and programme code generators, produce vast amounts of new content. But who owns these texts, images or scripts? Who holds the copyright to the products of GPT systems? And how is the handling of sources regulated?

Legal status unclear

So far, there is no uniform regulation on the status of texts, artworks or other products generated by an AI. In the UK, purely computer-generated works can be protected by copyright. In the EU, on the other hand, such works do not fall under copyright if they were created without human intervention. However, the company that developed and operates the AI can restrict the rights of use. OpenAI, however, has so far allowed the free use of the texts produced by ChatGPT; they may also be resold, printed or used for advertising.

At first glance, this is clear and very practical for users. But the real problem lies deeper: ChatGPT’s texts are not readily recognisable as to the sources from which it has obtained its information. Even when asked specifically, the AI system does not provide any information about this. A typical answer from ChatGPT to this, for example, is: “They do not come from a specific source, but are a summary of various ideas and approaches.”

The problem of training data

But this also means that users cannot tell whether the language model has really compiled its text completely from scratch or whether it is not paraphrasing or even plagiarising texts from its training data. Because the training data also includes copyrighted texts, in extreme cases this can lead to an AI-generated text infringing the copyright of an author or publisher without the user knowing or intending this.

Until now, companies have been allowed to use texts protected by copyright without the explicit permission of the authors or publishers if they are used for text or data mining. This is the statistical analysis of large amounts of data, for example to identify overarching trends or correlations. Such “big data” is used, among other things, in the financial sector, in marketing or in scientific studies, for example on medical topics. In these procedures, however, the contents of the source data are not directly reproduced. This is different with GPT systems.

Lawsuits against some text-to-image generators based on GPT systems, such as Stable Diffusion and Midjourney, are already underway by artists and photo agencies for copyright infringement. The AI systems had used part of protected artworks for their collages. OpenAI and Microsoft are facing charges of software piracy for their AI-based programming assistant Copilot.

Are ChatGPT and Co. plagiarising?

Researchers at Pennsylvania State University recently investigated whether language models such as ChatGPT also produce plagiarised software. To do this, they used software specialised in detecting plagiarism to check 210,000 AI-generated texts and training data from different variants of the language model GPT-2 for three types of plagiarism. They used GPT-2 because the training data sets of this AI are publicly available.

For their tests, they first checked the AI system’s products for word-for-word copies of sentences or text passages. Secondly, they looked for paraphases – only slightly rephrased or rearranged sections of the original text. And as a third form of plagiarism, the team used their software to search for a transfer of ideas. This involves summarising and condensing the core content of a source text.

From literal adoption to idea theft

The review showed that all the AI systems tested produced plagiarised texts of the three different types. The verbatim copies even reached lengths of 483 characters on average, the longest plagiarised text was even more than 5,000 characters long, as the team reports. The proportion of verbatim plagiarism varied between 0.5 and almost 1.5 per cent, depending on the language model. Paraphrased sections, on the other hand, averaged less than 0.5 per cent.

Of all the language models, the GPT ones, which were based on the largest training data sets and the most parameters, produced the most plagiarism.

“The larger a language model is, the greater its abilities usually are,”

Jooyoung Lee

explains first author Jooyoung Lee. “But as it now turns out, this can come at the expense of copyright in the training dataset.” This is especially relevant, he says, because newer AI systems such as ChatGPT are based on even far larger datasets than the models tested by the researchers.

“Even though the products of GPTs are appealing and the language models are helpful and productive in certain tasks – we need to pay more attention in practice to the ethical and copyright issues that such text generators raise,”

Thai Lee

says co-author Thai Le from the University of Mississippi.

Legal questions open

Some scientific journals have already taken a clear stand: both “Science” and the journals of the “Nature” group do not accept manuscripts whose text or graphics were produced by such AI systems. ChatGPT and co. may also not be named as co-authors. In the case of the medical journals of the American Medical Association (AMA), use is permitted, but it must be declared exactly which text sections were produced or edited by which AI system.

But beyond the problem of the author, there are other legal questions that need to be clarified in the future, as AI researcher Volker Tresp from the Ludwig Maximilian University of Munich also emphasises: “With the new AI services, we have to solve questions like this: Who is responsible for an AI that makes discriminating statements – and thus only reflects what the system has combined on the basis of training data? Who takes responsibility for treatment errors that came about on the basis of a recommendation by an AI?” So far, there are no or only insufficient answers to these questions.

24 February 2023 – Author: Nadja Podbregar – published in German on www.scinexx.de