Question A:
The European Union's AI Act was approved by the European Parliament in March 2024: https://artificialintelligenceact.eu/the-act/
The EU's legislation to regulate artificial intelligence is likely to put European technology firms at a substantial disadvantage to their competitors elsewhere in the world.
Responses
Responses weighted by each expert's confidence
Question B:
By providing a clear set of rules, the EU's legislation on artificial intelligence is likely to enhance research and innovation by firms building the new technology.
Responses
Responses weighted by each expert's confidence
Question A Participant Responses
| Participant | University | Vote | Confidence | Bio/Vote History |
|---|---|---|---|---|
 Franklin Allen |
Imperial College London | Bio/Vote History | ||
|
Location is not particularly important when it comes to developing new types of AI so if it is easier to do in one place than another, that's where it will occur.
|
||||
 Pol Antras |
Harvard | Bio/Vote History | ||
|
|
||||
 Olivier Blanchard |
Peterson Institute | Bio/Vote History | ||
|
having read (part) of the regulation: the complexity of the tasks needed to satisfy the regulation may lead some entrepreneurs to move outside the EU.
|
||||
 Richard William Blundell |
University College London | Did Not Answer | Bio/Vote History | |
|
|
||||
 Maristella Botticini |
Bocconi | Bio/Vote History | ||
|
|
||||
 Agnès Bénassy-Quéré |
Paris School of Economics | Bio/Vote History | ||
|
Maybe more difficult in the short term but more sustainable in the medium to long term, which raises the issue of hysteresis. Copyright legislation is likely to be a key issue.
|
||||
 Elena Carletti |
Bocconi | Did Not Answer | Bio/Vote History | |
|
|
||||
 Jean-Pierre Danthine |
Paris School of Economics | Did Not Answer | Bio/Vote History | |
|
|
||||
 Paul De Grauwe |
LSE | Bio/Vote History | ||
|
|
||||
 Jan Eeckhout |
UPF Barcelona | Bio/Vote History | ||
|
Following the "California effect" (tech firms adopt CA regulation for the entire US or even worldwide), there is likely to be the "Brussels effect": tech firms may well adopt EU regulation globally
|
||||
 Xavier Freixas |
Barcelona GSE | Did Not Answer | Bio/Vote History | |
|
|
||||
 Nicola Fuchs-Schündeln |
Goethe-Universität Frankfurt | Bio/Vote History | ||
|
|
||||
 Jordi Galí |
Barcelona GSE | Bio/Vote History | ||
|
|
||||
 Luis Garicano |
LSE | Bio/Vote History | ||
|
The lAI act certainly increases compliance costs. This is a minus. But it also provides incentives for open source and transparency, which could facilitate catch up. The balance is probably negative.
-see background information here |
||||
 Yuriy Gorodnichenko |
Berkeley | Did Not Answer | Bio/Vote History | |
|
|
||||
 Rachel Griffith |
University of Manchester | Bio/Vote History | ||
|
|
||||
 Veronica Guerrieri |
Chicago Booth | Bio/Vote History | ||
|
|
||||
 Luigi Guiso |
Einaudi Institute for Economics and Finance | Bio/Vote History | ||
|
|
||||
 Sergei Guriev |
Sciences Po | Bio/Vote History | ||
|
|
||||
 Patrick Honohan |
Trinity College Dublin | Bio/Vote History | ||
|
Is this what will hold them back? I doubt it.
|
||||
 Beata Javorcik |
University of Oxford | Bio/Vote History | ||
|
|
||||
 Jan Pieter Krahnen |
Goethe University Frankfurt | Bio/Vote History | ||
|
The rules of the AI Act apply to all developers who intend to become active in Europe, no matter where they are located and whether they are final or intermediate product.
|
||||
 Botond Kőszegi |
Central European University | Did Not Answer | Bio/Vote History | |
|
|
||||
 Eliana La Ferrara |
Harvard Kennedy | Did Not Answer | Bio/Vote History | |
|
|
||||
 Christian Leuz |
Chicago Booth | Bio/Vote History | ||
|
Constraining EU firms when competitors are unconstrained is a disadvantage. But how substantial it is depends also on what other countries do, hence uncertain. Even if EU firms are disadvantaged, restrictions can be right if activity has substantial externalities or risks.
|
||||
 Thierry Mayer |
Sciences-Po | Bio/Vote History | ||
|
|
||||
 Costas Meghir |
Yale | Bio/Vote History | ||
|
|
||||
 Marco Pagano |
Università di Napoli Federico II | Bio/Vote History | ||
|
|
||||
 Lubos Pastor |
Chicago Booth | Bio/Vote History | ||
|
|
||||
 Torsten Persson |
Stockholm University | Did Not Answer | Bio/Vote History | |
|
|
||||
 Christopher Pissarides |
London School of Economics and Political Science | Did Not Answer | Bio/Vote History | |
|
|
||||
 Richard Portes |
London Business School | Bio/Vote History | ||
|
|
||||
 Canice Prendergast |
Chicago Booth | Bio/Vote History | ||
|
|
||||
 Carol Propper |
Imperial College London | Did Not Answer | Bio/Vote History | |
|
|
||||
 Imran Rasul |
University College London | Did Not Answer | Bio/Vote History | |
|
|
||||
 Lucrezia Reichlin |
London Business School | Did Not Answer | Bio/Vote History | |
|
|
||||
 Ricardo Reis |
London School of Economics | Did Not Answer | Bio/Vote History | |
|
|
||||
 Rafael Repullo |
CEMFI | Bio/Vote History | ||
|
|
||||
 Hélène Rey |
London Business School | Did Not Answer | Bio/Vote History | |
|
|
||||
 Antoinette Schoar |
MIT | Did Not Answer | Bio/Vote History | |
|
|
||||
 Kjetil Storesletten |
University of Minnesota | Bio/Vote History | ||
|
|
||||
 Daniel Sturm |
London School of Economics | Bio/Vote History | ||
|
|
||||
 Silvana Tenreyro |
LSE | Did Not Answer | Bio/Vote History | |
|
|
||||
 John Van Reenen |
LSE | Bio/Vote History | ||
|
|
||||
 Rick Van der Ploeg |
Oxford | Bio/Vote History | ||
|
|
||||
 John Vickers |
Oxford | Bio/Vote History | ||
|
|
||||
 Hans-Joachim Voth |
University of Zurich | Bio/Vote History | ||
|
|
||||
 Karl Whelan |
University College Dublin | Did Not Answer | Bio/Vote History | |
|
|
||||
 Charles Wyplosz |
The Graduate Institute Geneva | Bio/Vote History | ||
|
|
||||
Question B Participant Responses
| Participant | University | Vote | Confidence | Bio/Vote History |
|---|---|---|---|---|
|
Franklin Allen |
Imperial College London | Bio/Vote History | ||
|
It's very difficult to write a clear set of rules for something that is novel and as dynamic as AI.
|
||||
|
Pol Antras |
Harvard | Bio/Vote History | ||
|
|
||||
|
Olivier Blanchard |
Peterson Institute | Bio/Vote History | ||
|
this is the other side of the coin. a measure with costs and benefits...
|
||||
|
Richard William Blundell |
University College London | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Maristella Botticini |
Bocconi | Bio/Vote History | ||
|
|
||||
|
Agnès Bénassy-Quéré |
Paris School of Economics | Bio/Vote History | ||
|
|
||||
|
Elena Carletti |
Bocconi | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Jean-Pierre Danthine |
Paris School of Economics | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Paul De Grauwe |
LSE | Bio/Vote History | ||
|
|
||||
|
Jan Eeckhout |
UPF Barcelona | Bio/Vote History | ||
|
A huge risk for investment in AI is policy risk. Delayed or haphazard regulation is worse for the tech industry than the EU AI act
|
||||
|
Xavier Freixas |
Barcelona GSE | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Nicola Fuchs-Schündeln |
Goethe-Universität Frankfurt | Bio/Vote History | ||
|
|
||||
|
Jordi Galí |
Barcelona GSE | Bio/Vote History | ||
|
|
||||
|
Luis Garicano |
LSE | Bio/Vote History | ||
|
There is little question that, in areas defined as harmful, including e.g. cars and large LLMs (size is a measure of risk in the legislation for GenAI models) the regulatory and compliance burden is larger and hence research and innovation will decrease.
|
||||
|
Yuriy Gorodnichenko |
Berkeley | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Rachel Griffith |
University of Manchester | Bio/Vote History | ||
|
|
||||
|
Veronica Guerrieri |
Chicago Booth | Bio/Vote History | ||
|
|
||||
|
Luigi Guiso |
Einaudi Institute for Economics and Finance | Bio/Vote History | ||
|
|
||||
|
Sergei Guriev |
Sciences Po | Bio/Vote History | ||
|
|
||||
|
Patrick Honohan |
Trinity College Dublin | Bio/Vote History | ||
|
It's not evident that the rules are yet clear
|
||||
|
Beata Javorcik |
University of Oxford | Bio/Vote History | ||
|
|
||||
|
Jan Pieter Krahnen |
Goethe University Frankfurt | Bio/Vote History | ||
|
Other countries, nota by US, may follow suit with similar rules at some point. So, following these rules will likely be the rational strategy for all AI companies - which means that European developers are not at a distinct disadvantage or an advantage either.
|
||||
|
Botond Kőszegi |
Central European University | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Eliana La Ferrara |
Harvard Kennedy | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Christian Leuz |
Chicago Booth | Bio/Vote History | ||
|
Providing a clear set of rules removes regulatory uncertainty, which should promote development of AI systems, but only of those that fall within the rules. It also constrains certain developments & innovations, which again can be appropriate depending on externalities and risks
|
||||
|
Thierry Mayer |
Sciences-Po | Bio/Vote History | ||
|
|
||||
|
Costas Meghir |
Yale | Bio/Vote History | ||
|
|
||||
|
Marco Pagano |
Università di Napoli Federico II | Bio/Vote History | ||
|
|
||||
|
Lubos Pastor |
Chicago Booth | Bio/Vote History | ||
|
|
||||
|
Torsten Persson |
Stockholm University | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Christopher Pissarides |
London School of Economics and Political Science | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Richard Portes |
London Business School | Bio/Vote History | ||
|
|
||||
|
Canice Prendergast |
Chicago Booth | Bio/Vote History | ||
|
|
||||
|
Carol Propper |
Imperial College London | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Imran Rasul |
University College London | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Lucrezia Reichlin |
London Business School | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Ricardo Reis |
London School of Economics | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Rafael Repullo |
CEMFI | Bio/Vote History | ||
|
|
||||
|
Hélène Rey |
London Business School | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Antoinette Schoar |
MIT | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Kjetil Storesletten |
University of Minnesota | Bio/Vote History | ||
|
|
||||
|
Daniel Sturm |
London School of Economics | Bio/Vote History | ||
|
|
||||
|
Silvana Tenreyro |
LSE | Did Not Answer | Bio/Vote History | |
|
|
||||
|
John Van Reenen |
LSE | Bio/Vote History | ||
|
|
||||
|
Rick Van der Ploeg |
Oxford | Bio/Vote History | ||
|
|
||||
|
John Vickers |
Oxford | Bio/Vote History | ||
|
|
||||
|
Hans-Joachim Voth |
University of Zurich | Bio/Vote History | ||
|
|
||||
|
Karl Whelan |
University College Dublin | Did Not Answer | Bio/Vote History | |
|
|
||||
|
Charles Wyplosz |
The Graduate Institute Geneva | Bio/Vote History | ||
|
|
||||
