Delivery should be verified.
1 Introduction
Ladies and gentlemen,
The science fiction writer William Gibson once said: The future is already here – it is just not very evenly distributed
.[1] That remark accurately describes the current state of artificial intelligence. Its capabilities are already visible across many applications – from text generation and coding to research and forecasting. Or, in a more personal context, from designing invitation cards to making songs or even movies. However, its broader economic impact is still not fully evident from the aggregate statistics.[2]
This is not without historical precedent. When electricity first spread throughout the advanced economies, the productivity gains were limited initially. Firms adopted electric motors, but they did not reorganize their production right away.[3] However, artificial intelligence could be spreading much faster than previous general-purpose technologies.[4] This indicates that artificial intelligence is a transformation that will likely have a significant impact on the global economy.
In my speech today, I will explore what AI may mean for growth, inflation, and financial stability. Another key question I will address is this: How do we shape its impact? In this context, I would like to discuss how well Europe is positioned in the global AI race. But before I do that, let me outline how I view AI and where I see its strengths and weaknesses.
2 What today’s AI can and cannot do
AI is best understood not as a single product, but as a general-purpose technology. Such technologies are widely used across many sectors. They continuously improve over time and interact with other innovations.[5] As they evolve, they spread throughout the economy and generate broad productivity gains. Economic history provides several prominent examples. The steam engine, for instance, did not just enhance one industry; it fundamentally transformed transportation, manufacturing, and mining, becoming a cornerstone of the Industrial Revolution. Electricity followed a similar trajectory: once widely adopted, it enabled entirely new production processes, reshaped factory organization, and ultimately powered the mass production systems of the 20th century. With a delay of about 140 years, electricity is now finally making strides in individual transportation. More recently, digital computing and the internet have also been general-purpose technologies, driving innovations from industrial automation to global financial systems.
Artificial intelligence shares these defining characteristics. It is not limited to a single use case but can be applied across almost all sectors – from healthcare and finance to manufacturing and public administration. In this context, AI can be described as “the steam engine of the mind”[6]. The steam engine amplified human physical labor. AI amplifies human cognitive abilities. When used effectively, it will enhance human intelligence, potentially leading to significant impacts on productivity, innovation, and economic growth.
The strengths of AI are becoming increasingly evident: It can process and summarize vast amounts of text; detect patterns in large datasets; support coding and scientific research; improve forecasting in certain settings; and automate repetitive knowledge tasks at high speed. However, the weaknesses of AI are equally important for users, as generative AI can produce false or biased outputs, hallucinate, and reproduce errors hidden in training data. These limitations stem from the nature of current AI systems, which are based on statistical methods and generate responses probabilistically. They cannot reason in a human sense but predict likely word sequences based on patterns in their training data.
Another significant drawback of AI is the substantial amount of electricity it consumes.[7] The energy consumption of AI can be illustrated by the heat generated by data centers. Estimates suggest that land surface temperatures near AI data centers can increase by approximately 2°C after operations commence[8]. This can create local microclimates, known as the «data heat island effect,» which may have tangible effects on nearby communities. In total, over 340 million people worldwide could be exposed to such local warming effects.
Overall, the AI boom is expected to lead to a significant increase in electricity demand in the coming years. Currently, data centers consume around 415 terawatt-hours of electricity annually, roughly 1.5% of global electricity consumption in 2024. The International Energy Agency projects that global electricity consumption for data centers will double by 2030, representing just under 3% of total global electricity consumption at that time.[9] One might say: As the «steam engine of the mind,» AI comes with a significant electricity cost.
3 AI and the macroeconomy: growth, investment, and emerging risks
These developments highlight that AI is not just a technological transformation but also an economic one with far-reaching implications. These include a restructuring of the labor market, the emergence of new industries, and shifts in global trade patterns. While these aspects are crucial, I will focus on the implications of artificial intelligence for growth, inflation, and financial stability.
Starting with economic growth, productivity growth is the primary channel through which technological progress translates into economic growth. However, past technological innovations typically increased productivity only after a significant delay as they required widespread adoption, complementary investments, and organizational changes, which took time. Existing capital and practices persist, so new technologies are initially added to old systems rather than completely replacing them. For instance, electric power took several decades to fully impact productivity[11]. Computers are another example.
The productivity boom driven by IT occurred in the late 1990s, more than two decades after personal computers were introduced. However, the adoption of AI could be much faster, with over 1.2 billion people worldwide using AI tools within three years, surpassing the adoption rates of the internet, smartphones, and personal computers. This rapid adoption of AI is due to building on existing digital infrastructure and speeding up the innovation process itself.
Estimates suggest that rapid adoption of AI could elevate annual labor productivity growth by 0.8 to 1.3 percentage points in G7 economies over the next decade, compared to the slower productivity growth seen in recent years. The impact of AI on productivity remains uncertain and depends on factors such as the pace of adoption among firms.
In the long run, AI-related productivity gains could offset losses from demographic decline in the working population and significantly increase aggregate output, consumption, and investment. AI investment is already contributing to GDP growth, surpassing the contribution of IT components during the dot-com boom.
The potential effects of AI on inflation are uncertain, with higher productivity initially dampening inflationary pressures but increased demand over time potentially leading to inflation. AI adoption may also affect pricing strategies and raise concerns about financial stability risks due to AI-supplier concentration, herding behavior, and cyber and operational risks.
The use of AI in lending decisions by banks could lead to more closely aligned credit risk assessments and amplify procyclical dynamics, making credit availability more sensitive to the business cycle. A strong reliance on a limited number of AI providers could also pose operational and data protection risks. Additionally, the use of AI in the financial sector introduces new cyber risks due to autonomous AI agents potentially exhibiting harmful behavior. Timely identification and mitigation of these risks are essential for maintaining financial stability, especially in light of the ongoing discussion surrounding Anthropic’s Mythos. Mythos, an AI model, has the potential to swiftly detect and exploit security vulnerabilities in financial institutions’ software. However, while this technology can enhance digital security systems, it also poses a risk of exploitation for malicious purposes. It is crucial to prevent the misuse of this technology and ensure that all relevant institutions have access to it to prevent any competitive distortions.
Moving forward, it is important to acknowledge the broader structural issue of the concentrated and interconnected development and deployment of AI globally. This underscores the significance of assessing the standing of different regions in this evolving landscape, particularly Europe’s position in the global AI race. Despite entering the AI era with strengths, Europe lags behind the United States and China in various aspects.
The United States leads in frontier-model development and private AI investment, with China following suit. In contrast, Europe falls behind in AI investment, scale-up financing, computing infrastructure, and frontier model development. However, Europe does possess substantial research potential that can fuel entrepreneurial endeavors. Noteworthy private suppliers of AI services in Europe, such as Mistral AI, Black Forest Labs, and Aleph Alpha, demonstrate the continent’s capabilities.
Collaboration among central banks, like the Bundesbank and Banca d’Italia, has proven beneficial in monitoring AI adoption trends in Italy and Germany. Central banks are actively involved in shaping the AI transformation within their institutions, with the Bundesbank implementing a comprehensive AI strategy. To fully harness the potential of AI, progress at the European level is imperative. Initiatives like the EU’s AI Act and InvestAI aim to accelerate AI development in Europe.
As the use of AI grows, the demand for reliable and affordable energy to power AI infrastructure becomes critical. This challenge is not unique to Europe and underscores the importance of addressing energy needs alongside capital mobilization for AI advancement. In the United States, around half of the planned data centre projects have been delayed or cancelled, with power infrastructure constraints being a significant factor. Similarly, in Europe, new data centre projects have faced delays or postponements due to insufficient grid capacity. For example, in Dublin and Frankfurt, it can take three to five years to supply power to new data centres. This poses a risk of investments not fully materializing. To address these challenges, the European Union is developing a policy framework, including the Data Centre Energy Efficiency Package expected in 2026. These developments highlight the importance of having the right institutional frameworks and effective implementation to successfully deploy artificial intelligence. (2026), Lernen, Anwenden, Gestalten: Die Verknüpfung von Theorie und Praxis als Erfolgsrezept in einer sich wandelnden Arbeitswelt, Festvortrag an der DHBW Karlsruhe, 17. März 2026.
in german while maintaining its SEO structure and link types (dofollow or nofollow) exactly as they appear in
Check against delivery.
1 Einführung
Damen und Herren,
Der Science-Fiction-Autor William Gibson sagte einmal: «Die Zukunft ist bereits hier – sie ist nur noch nicht sehr gleichmäßig verteilt».[1] Diese Bemerkung beschreibt recht treffend den aktuellen Stand der künstlichen Intelligenz. Deren Fähigkeiten sind bereits in vielen Anwendungen sichtbar – von Textgenerierung und Codierung über Forschung und Prognosen. Oder, in einem persönlicheren Kontext, von der Gestaltung von Einladungskarten bis hin zur Erstellung von Liedern oder sogar Filmen. Dennoch ist ihre breitere wirtschaftliche Wirkung aus den aggregierten Statistiken noch weit weniger ersichtlich.[2]
Dies ist nicht ohne historischen Präzedenzfall. Als sich die Elektrizität erstmals in den fortgeschrittenen Volkswirtschaften verbreitete, waren die Produktivitätsgewinne zunächst begrenzt. Unternehmen übernahmen elektrische Motoren, aber sie organisierten ihre Produktion nicht sofort neu.[3] Die künstliche Intelligenz könnte jedoch erheblich schneller verbreitet sein als frühere General Purpose Technologies.[4] Dies legt nahe, dass künstliche Intelligenz eine Transformation ist, die mit großer Wahrscheinlichkeit einen massiven Einfluss auf die globale Wirtschaft haben wird.
In meiner heutigen Rede werde ich untersuchen, was KI für Wachstum, Inflation und finanzielle Stabilität bedeuten kann. Eine weitere wichtige Frage, der ich nachgehen werde, ist: Wie gestalten wir ihre Auswirkungen? In diesem Zusammenhang möchte ich darauf eingehen, wie gut Europa im globalen KI-Wettbewerb positioniert ist. Doch bevor ich das tue, erlauben Sie mir zu umreißen, wie ich KI sehe und wo ich ihre Stärken und Schwächen sehe.
2 Was die heutige KI kann und nicht kann
Künstliche Intelligenz wird am besten nicht als ein einzelnes Produkt, sondern als eine General Purpose Technology verstanden. Solche Technologien werden in vielen Sektoren weit verbreitet eingesetzt. Sie verbessern sich kontinuierlich im Laufe der Zeit. Und sie interagieren mit anderen Innovationen.[5] Bei ihrer Entwicklung verbreiten sie sich in der gesamten Wirtschaft und generieren breite Produktivitätsgewinne. Die Wirtschaftsgeschichte bietet mehrere prominente Beispiele. Die Dampfmaschine verbesserte beispielsweise nicht nur eine Branche; sie transformierte grundlegend Transport, Fertigung und Bergbau und wurde zum Eckpfeiler der Industriellen Revolution. Die Elektrizität folgte einem ähnlichen Weg: einmal weit verbreitet, ermöglichte sie völlig neue Produktionsprozesse, formte die Organisation von Fabriken um und versorgte letztendlich die Massenproduktionssysteme des 20. Jahrhunderts. Mit einer Verzögerung von etwa 140 Jahren marschiert die Elektrizität nun endlich zum Sieg im individuellen Transport. In jüngerer Zeit waren auch digitale Rechner und das Internet General Purpose Technologies, die Innovationen von der industriellen Automation bis zu globalen Finanzsystemen antrieben.
Künstliche Intelligenz teilt diese charakteristischen Merkmale. Sie ist nicht auf einen einzigen Anwendungsfall beschränkt. Sie kann in fast allen Branchen eingesetzt werden – von Gesundheitswesen und Finanzen über Fertigung und öffentliche Verwaltung. Vor diesem Hintergrund kann KI als «die Dampfmaschine des Geistes» bezeichnet werden.[6] Die Dampfmaschine verstärkte menschliche körperliche Arbeit. KI verstärkt menschliche kognitive Fähigkeiten. Richtig eingesetzt wird sie die menschliche Intelligenz verbessern – mit möglicherweise weitreichenden Auswirkungen auf Produktivität, Innovation und wirtschaftliches Wachstum.
In diesem Zusammenhang werden die Stärken der KI immer deutlicher: Sie kann riesige Textmengen verarbeiten und zusammenfassen; Muster in großen Datensätzen erkennen; Codierung und wissenschaftliche Forschung unterstützen; Prognosen in einigen Bereichen verbessern; und repetitive Wissensaufgaben in sehr hoher Geschwindigkeit automatisieren. Gleichzeitig sind die Schwächen der KI für die Nutzer ebenso wichtig, da generative KI halluzinieren kann; falsche oder voreingenommene Ergebnisse produzieren; und Fehler reproduzieren, die in den Trainingsdaten versteckt sind. Diese Einschränkungen spiegeln die zugrunde liegende Natur der aktuellen KI-Systeme wider. Große Sprachmodelle basieren auf statistischen Methoden und generieren Antworten probabilistisch. Sie können nicht im menschlichen Sinne argumentieren – vielmehr sagen sie wahrscheinliche Wortfolgen basierend auf Mustern in den Daten vorher, auf denen sie trainiert wurden.
Ein weiterer wichtiger Nachteil von KI ist der enorme Energieverbrauch, den sie verursacht.[7] Wie viel Energie KI verbraucht, lässt sich anhand der von Rechenzentren erzeugten Wärme veranschaulichen. Schätzungen zeigen, dass die Landtemperaturen in Gebieten rund um KI-Rechenzentren nach Inbetriebnahme um etwa 2°C steigen können.[8] Dies kann lokale Mikroklimata erzeugen, die oft als «Data Heat Island Effect» bezeichnet werden. Diese Temperaturerhöhungen können spürbare Auswirkungen auf nahegelegene Gemeinden haben. Insgesamt könnten weltweit mehr als 340 Millionen Menschen solchen lokalen Erwärmungseffekten ausgesetzt sein.
Insgesamt wird der KI-Boom voraussichtlich in den kommenden Jahren einen dramatischen Anstieg des Strombedarfs auslösen. Derzeit verbrauchen Rechenzentren etwa 415 Terawattstunden Strom pro Jahr. This amounts to approximately 1.5% of global electricity consumption in 2024. The International Energy Agency predicts that by 2030, data centers will consume double the amount of global electricity, making up nearly 3% of total consumption. One could say that as the «steam engine of the mind,» AI comes with a significant electricity cost. The use of artificial intelligence (AI) may initially lead to inflationary effects, particularly as algorithms could enable prices to be set above competitive levels without coordination. This poses a risk from a central banking perspective, requiring careful monitoring. Additionally, AI could contribute to financial stability risks through concentration among AI suppliers, herding behavior, market correlation, and cyber and operational vulnerabilities. For example, reliance on AI models for credit assessments could lead to synchronized credit curtailment across institutions in response to economic downturns, increasing financial stability risks. The deployment of AI in the financial sector also introduces sophisticated cyber risks, necessitating early identification and mitigation efforts to safeguard against potential threats.
Moreover, the concentration of AI development and deployment globally underscores the importance of assessing different regions’ positions in the evolving landscape. Europe, while entering the AI era with notable strengths, lags behind the United States and China in terms of frontier-model development and private investment. However, Europe’s research potential and existing private suppliers of AI services offer opportunities for entrepreneurial activity and innovation. Collaborative efforts among central banks, such as the Bundesbank and Banca d’Italia, are crucial in addressing these challenges and leveraging AI technologies for economic growth and stability. In 2024 and 2025, companies in Italy and Germany were surveyed with a standardized set of questions regarding the use of generative artificial intelligence (AI). The findings revealed a significant increase in AI adoption in both countries.
Central banks, including the Bundesbank, are actively involved in shaping the transformation brought about by AI within their institutions. The Bundesbank has approved a comprehensive strategy for AI, with two out of three colleagues utilizing AI regularly in a safe environment. The primary goal is to systematically integrate AI into the Bundesbank’s operations to enhance task fulfillment.
While individual efforts by firms and public institutions are important, progress at the European level is necessary to realize the full potential of AI. Initiatives such as the AI Act and InvestAI in the European Union are steps towards financing, scaling, infrastructure, skills, energy, and market integration for AI development.
The availability of reliable and affordable energy is crucial for running AI infrastructure effectively. Challenges in power infrastructure, both in Europe and the United States, have led to delays in data center projects. The EU is working towards a comprehensive policy framework to ensure that AI infrastructure expansion aligns with energy and climate goals, with initiatives like the Data Centre Energy Efficiency Package expected in 2026.
In conclusion, the potential of AI to impact the world economy is significant, but effective adoption across sectors and adaptation by firms and institutions are crucial. Policymakers and central banks play a role in scrutinizing AI’s effects on inflation and financial stability, supporting capital and computing power for AI development, and ensuring equitable distribution of AI benefits across economies. Europe possesses the necessary elements to master the AI revolution and should focus on innovation, investment, institutional frameworks, and practical implementation.
1 Introducción
Señoras y señores,
El escritor de ciencia ficción William Gibson dijo una vez: El futuro ya está aquí, solo que no está muy distribuido equitativamente
.[1] Esa observación describe bastante bien el estado actual de la inteligencia artificial. Sus capacidades ya son visibles en muchas aplicaciones, desde la generación de texto y programación hasta la investigación y pronósticos. O, en un contexto más personal, desde el diseño de tarjetas de invitación hasta la creación de canciones o incluso películas. Sin embargo, su impacto económico más amplio aún no es muy evidente en las estadísticas agregadas.[2]
Esto no carece de precedentes históricos. Cuando la electricidad se extendió por primera vez en las economías avanzadas, los incrementos de productividad fueron limitados al principio. Las empresas adoptaron motores eléctricos, pero no reorganizaron su producción de inmediato.[3] Sin embargo, la inteligencia artificial podría estar propagándose significativamente más rápido que las tecnologías generales anteriores. Esto sugiere que la inteligencia artificial es una transformación que casi con seguridad tendrá un gran impacto en la economía global.
En mi discurso de hoy, exploraré lo que la inteligencia artificial podría implicar para el crecimiento, la inflación y la estabilidad financiera. Otra pregunta clave que abordaré es: ¿cómo moldeamos su impacto? En este contexto, me gustaría abordar qué tan bien está posicionada Europa en la carrera mundial de la inteligencia artificial. Pero antes de hacerlo, permítanme esbozar cómo veo la inteligencia artificial y dónde veo sus fortalezas y debilidades.
2 Lo que la inteligencia artificial de hoy puede y no puede hacer
La inteligencia artificial se entiende mejor no como un único producto, sino como una tecnología de propósito general. Estas tecnologías se utilizan ampliamente en muchos sectores. Mejoran continuamente con el tiempo. E interactúan con otras innovaciones.[5] A medida que evolucionan, se extienden por toda la economía y generan amplios incrementos de productividad. La historia económica ofrece varios ejemplos prominentes. Por ejemplo, la máquina de vapor no solo mejoró una industria; transformó fundamentalmente el transporte, la manufactura y la minería, convirtiéndose en un pilar de la Revolución Industrial. La electricidad siguió una trayectoria similar: una vez ampliamente adoptada, permitió procesos de producción completamente nuevos, reorganizó la organización de fábricas y finalmente impulsó los sistemas de producción en masa del siglo XX. Con un retraso de aproximadamente 140 años, la electricidad ahora avanza triunfalmente hacia la victoria en el transporte individual. Más recientemente, la computación digital y el internet también han sido tecnologías de propósito general, impulsando innovaciones que van desde la automatización industrial hasta los sistemas financieros globales.
La inteligencia artificial comparte estas características definitorias. No está limitada a un solo caso de uso. Se puede utilizar en casi todos los sectores, desde la salud y las finanzas hasta la manufactura y la administración pública. En este contexto, la inteligencia artificial se puede describir como «la máquina de vapor de la mente».[6] La máquina de vapor amplificó el trabajo físico humano. La inteligencia artificial amplifica las capacidades cognitivas humanas.
When utilized correctly, artificial intelligence has the potential to improve human intelligence, leading to increased productivity, innovation, and economic growth. AI can process large amounts of text, detect patterns in data, support coding and research, improve forecasting, and automate repetitive tasks. However, AI also has limitations, such as generating false outputs and consuming a significant amount of electricity. Despite these drawbacks, the adoption of AI is rapidly increasing, with the potential to significantly boost productivity growth in the future. This could help offset losses from demographic decline and lead to higher output, consumption, and investment. The impact of AI is expected to continue growing and shaping the economy in the years to come. In the US, investments related to artificial intelligence (AI) are already making a noticeable contribution to GDP growth. By 2025, AI investments are projected to contribute around 1 percentage point to GDP growth, surpassing the contribution of IT components during the dot-com boom. In Germany, firm expenditure on generative AI is already comparable to traditional digital investments, with most of the spending going towards recurring costs such as subscriptions and IT staff.
The potential impact of AI on inflation is uncertain. While higher productivity from AI adoption may initially reduce inflationary pressures by lowering firms’ costs and addressing labor force shortages, over time it could lead to increased demand for electricity and other resources, potentially fueling inflation. The use of algorithms in pricing could also lead to setting prices above competitive levels, further complicating the inflation outlook.
From a financial stability perspective, the concentration of AI development and deployment poses risks such as supplier concentration, herding behavior, and cyber risks. The use of AI in lending decisions could lead to increased homogeneity in credit risk assessments across institutions, amplifying procyclical dynamics and increasing financial stability risks.
In Europe, AI development is gaining momentum but still lags behind the US and China in terms of investment and frontier-model development. While the US leads in private investment and model development, China benefits from significant government investment in AI. Europe’s relative weakness is evident in scale-up financing, computing infrastructure, and frontier model development, areas dominated by American firms. Despite these challenges, Europe has immense research potential that can drive entrepreneurial activity in the AI sector. In Europe, we have a strong lineup of important private suppliers of AI services, such as Mistral AI, Black Forest Labs, and Aleph Alpha. Additionally, Europe is home to many hidden champions – small firms that are global leaders in their specific markets. The race to use AI in industrial processes is still open, and Europe, especially Germany, has a wealth of data for training highly specialized AI models. This enables the development of tailor-made AI solutions for various areas such as production, logistics, and maintenance.
The adoption of AI technologies among European firms is increasing, thanks to collaborations among central banks like the Bundesbank and Banca d’Italia. For example, a harmonized set of questions on the use of generative AI was posed to firms in Italy and Germany, revealing a marked increase in AI adoption in both countries.
Central banks are actively shaping the transformation by integrating AI into their operations. The Bundesbank has approved a comprehensive strategy for AI implementation, with two out of three colleagues using AI regularly in their work. Efforts at the individual and institutional levels are crucial, but progress at the European level is also necessary to realize the full potential of AI. Initiatives like the AI Act and InvestAI are steps in the right direction.
Availability of reliable and affordable energy is crucial for running AI infrastructure. Europe faces challenges with power infrastructure affecting data center projects. The EU is developing a policy framework to ensure AI infrastructure expansion aligns with energy and climate objectives.
In conclusion, AI has the potential to drive the world economy, but its impact depends on widespread adoption and adaptation by firms and institutions. Europe is well-equipped to master the AI revolution with its thriving AI sector, research capabilities, specialized corporations, and internal market. Active shaping of the AI revolution is essential, and policymakers can support this by providing the right policy mix for scale-up capital and computing power. F. and M. Trajtenberg (1995), General purpose technologies ‘Engines of growth’?, Journal of Econometrics, Vol. 65(1), pp. 83–108.
Hoffman, R., Gen AI: A cognitive industrial revolution, McKinsey, podcast hosted by Lareina Yee, 7 June 2024.
Bogmans et al. (2025), Power Hungry: How AI Will Drive Energy Demand, IMF Working Paper WP/25/81.
Marinoni et al. (2026), The data heat island effect: quantifying the impact of AI data centers in a warming world, mimeo.
International Energy Agency (2025), Energy and AI.
See David (1990).
See David (1990).
Oliner, S. D. and D. E. Sichel (2000), The Resurgence of Growth in the Late 1990s: Is Information Technology the Story? Journal of Economic Perspectives, 14(4), pp. 3–22.
AI Diffusion Report: Mapping Global AI Adoption and Innovation
Busch, M. and D. Duwe (2023), Artificial intelligence in innovation processes – A study using the example of an innovation research institute. Karlsruhe: Fraunhofer Institute for Systems and Innovation Research ISI.
Filippucci et al. (2025), Macroeconomic productivity gains from Artificial Intelligence in G7 economies, OECD Artificial Intelligence Paper No 41.
Goldin et al. (2024), Why Is Productivity Slowing Down? Journal of Economic Literature 62(1), pp. 196–268.
See Oliner and Sichel (2000).
See Acemoglu (2025), The simple macroeconomics of AI, Economic Policy 40 (121), pp. 13–58.
Deutsche Bundesbank (2026), Generative AI in German enterprises: adoption, costs and expected economic impact, Monthly Report, March 2026.
Asao et al. (2025), The Impact of Aging and AI on Japan’s Labor Market: Challenges and Opportunities, IMF Working Paper, No WP/25/184.
Aldasoro et al. (2024), The impact of artificial intelligence on output and inflation, BIS Working Paper No 1179.
Curran, E. and M. Niquette, AI-Led Investments Are Driving US Economic Growth, bloomberg.com, 31 October 2025.
Tracking AI’s Contribution to GDP Growth | St. Louis Fed
See Deutsche Bundesbank (2026).
Markman, J., Kevin Warsh’s New Playbook: AI, Productivity And A Deflation Bet, Forbes, 2 February 2026.
See Aldasoro et al. (2024).
E. Calavano et al. (2020). Artificial Intelligence, Algorithmic Pricing, and Collusion, American Economic Review 2020, 110(10).
Leitner et al., The rise of artificial intelligence: benefits and risks for financial stability, ECB Financial Stability Review 2024; IMF, Global Financial Stability Report 2024.
Stanford University, The 2025 AI Index Report, Chapter 4.
China nutzt KI als Schlüsseltechnologie | Special | China | KI-Strategie
Federal AI and IT Research and Development Spending Analysis | Federal Budget IQ; European approach to artificial intelligence | Shaping Europe’s digital future
Nagel, J. (2026), Lernen, Anwenden, Gestalten: Die Verknüpfung von Theorie und Praxis als Erfolgsrezept in einer sich wandelnden Arbeitswelt, keynote speech at the DHBW Karlsruhe, 17 March 2026.
Usage of AI technologies increasing in EU enterprises – News articles – Eurostat; See also CISCO (2026), State of Industrial AI, Report.
Bencivelli et al. (2026), Embracing AI in Europe: New evidence from harmonised central bank business surveys, VoxEU Column.
AI Act enters into force – European Commission
EU launches InvestAI initiative to mobilise €200 billion of investment in artificial intelligence | Shaping Europe’s digital future
Half of planned US data center builds have been delayed or canceled, growth limited by shortages of power infrastructure and parts from China – the AI build-out flips the breakers | Tom’s Hardware
Jacamon et al., Overcoming energy constraints is key to delivering on Europe’s data centre goals, International Energy Agency, 16 November 2025.
Granskog et al., The role of power in unlocking the European AI revolution, McKinsey, 24 October 2024.
Energy performance of data centres – Energy – European Commission Could you please rephrase this? Can you please rephrase that? Please rewrite the following sentence:
«The quick brown fox jumps over the lazy dog.»
Rewritten sentence:
«The fast brown fox leaps over the sluggish dog.» Please rewrite the sentence for me.
QUELLEN
