Rethinking Intellectual Property in the Era of Open Science and AI
Legend
Transcript markers
- text Sentence being played
- text Search result
- text Intervention cited by the link you opened
Click a sentence to play the video from that moment. Select text to quote it with the speaker, timestamp and link.
More options
Offset
Transcript offset
Export
Live capture
Versions of this video
Transcripts may differ slightly from one version to another.
STOA European Parliament
Katarina Miller
These rights are designed in different ways to encourage creativity, investment and of course innovation. However, scientific progress also depends on something else: sharing knowledge. And scientists build on previous discoveries, they reuse data, test hypothesis, challenge findings and collaborate across borders, so some of us also try to reproduce science. And this brings us to open science, which seeks to make research more accessible, transparent, reusable and collaborative. And at first glance, we seem to have a contradiction.
Intellectual property gives us a certain right to exclude, open science encourages us to include, and this is precisely what our speakers will help us explore today. So we begin with our Member of European Parliament Timo Wölken, who is also shadow rapporteur on the Legal Affairs Committee's report on copyright and generative AI. You will introduce the European policy challenges and opportunities of developing a responsible intellectual property framework in the age of AI, more or less. We will see. Followed then by our distinguished colleague Professor Richard Gold, who is connected online from Canada.
He is a distinguished James McGill Professor at McGill University in Canada and director of its Centre for Intellectual Property Policy. Richard will offer an international perspective and explore how IP and open science can act as two complementary drivers of innovation while raising important questions about access and technological sovereignty. Then we have from Sweden, Professor Gustaf Nelhans, associate professor of neuroscience at Karolinska Institutet and a leading contributor to the Horizon Europe IP4OpenScience project. Welcome. Gustaf will take us from policy to practice, examining how researchers and institutions can use IP strategically to support rather than obstruct open science.
And then we go from Sweden to Spain, where Professor Pere Puigdomènech, emeritus research professor at CSIC and member of ALLEA's Science Policy Standing Committee, will explore IP in new genomic techniques for crop development, where the balance between protecting inventions, enabling research, and ensuring access has direct implications for agriculture and society.
And our fifth speaker, and now we go back to Germany, is Annika Müller-Karabil, researcher at the University of Bremen and a co-developer of the Open Science Learning Gate. Annika, you will address a question that is sometimes overlooked: even with excellent policies and legal instruments, how can researchers actually apply them without the necessary knowledge, training, and infrastructures? And then finally, last but not least, we welcome Manuel Aleixo, cabinet expert to European Commissioner Iliana Ivanova, responsible for startups, research, and innovation. And Manuel will bring the European Commission's perspective and help us connect today's discussion to the broader ambition of building a European innovation system that is both sovereign and inclusive. And now, dear Timo Wölken, the floor is absolutely yours.
Tiemo Wölken
Hello everyone, also from my side.
I should be used to it, to activate the microphone before speaking now, but okay. So, dear colleagues, thank you very much for the warm welcome, dear Lina and especially dear Katharina. So I would like to first start obviously by thanking you for the invitation to let me speak on what I think is indeed a very important issue, because in the foundation it's about open science, intellectual property rights and AI. So what I would like to start with is a simple observation, and this is a bit continuing where Katharina just ended. So what I believe is that scientific progress is only possible because we can stand on the shoulders of scientific achievements of those who came before us.
So sharing is enormously important, but today those shoulders increasingly include also machines, and this is something we have to address.
Artificial intelligence is not new, but I think the that it is on the main stage is is new, but it is indeed reshaping the way we do science in fundamental ways. Researchers are increasingly relying on AI, politicians as well by the way, but that's not our focus today, but researchers heavily more and more rely on them to formulate questions, to evaluate their data, or even to write papers or sparing ideas. Briefly put, AI is, and I firmly believe that, a remarkable opportunity to supercharge our research, but on the other hand, obviously it comes with significant challenges. So it may make researchers less creative, less adventurous, and boxes in along already explored path of research. In other words, the accidental genius discovery may become less likely, at least at the current stage of AI.
This may obviously change with the progress, but for now I would say this is less, this will happen less. But apart from those rather philosophical challenges, AI also puts pressure on two pillars of our legal and scientific order: intellectual property rights and open science. And in this context, I want to discuss two specific questions today, briefly, which are two sides of the same coin from my perspective.
Firstly, in how far AI-generated content should benefit from IP protection, and secondly, how IP-protected content should be used for scientific research underpinned by AI. So let me begin with the first question. When an AI system generates a molecular structure or a research question or a block of analysis, who owns the result? Right now, I think we lack a clear answer, and I think this cannot go on. Universities, funders, companies need to know whether AI-generated results can be patented or copyrighted, or whether they should be open for all of us for use.
I believe in open science and the benefit of an open exchange for our common good. This may be naive, but still. Therefore, I am convinced that purely AI-generated output should not benefit from intellectual property protection, just like an AI-generated novel, artwork, or piece of music should not benefit from any exclusive rights. This position is based on a fundamental legal notion. IP protection exists to reward human creation, human creativity.
Copyright protects the author's creation, which is an expression of human choices, and patent law rewards human inventor, the person who advances the state of the art. Granting exclusive rights to purely machine-generated output would reward no one who deserves reward, while fencing off knowledge that could otherwise circulate freely. In this context, it is also important to note that virtually every AI system has been trained with freely available content, with the fruits of human creation, and it therefore follows logically that its output should also be freely available. But there's also a second danger. If AI output can be produced at industrial scale and then monopolized, we risk flooding the knowledge commons with exclusive claims, creating obstacles that burden researchers and entrench the power of those with the greatest computing power.
And I think this should not happen in Europe. However, there is an important distinction to make. What about research results that were generated by a human with the help or support of AI?
Where AI is simply a tool in the hands of a human, like a microscope or a hammer, the human's ingenuity remains the source of the creation and protection should remain available. A researcher who defines the question, uses the tool, then refines it, selects, edits, interrogates and takes responsibility for a result is the inventor or the author, whether they used a microscope or an AI system as a tool. So the dividing line I think we should pay attention to is human intellectual contribution. So the difficult nut to crack is the following: where exactly is that fine line between AI-generated and human-generated yet assisted content? In other words, what is the human creation threshold to become eligible for IP protection?
I cannot answer this highly complex question here today, but whatever the solution, I think that the EU should codify that line clearly so that the same scientific standards can be applied across the Union.
We have now discussed how AI output might or might not benefit from IP protection, and my second point concerns, but it's a rather short one, concerns the other side of the equation: the input. AI-driven research depends on analyzing vast quantities of existing material, protected material, articles, datasets, images, and text. And again, much of it, of it is protected by copyright. Europe, I think, made a wise choice here years ago in the Copyright Directive. The Directive established a text and data mining exemption for scientific research, allowing research organizations and cultural heritage institutions to mine lawfully accessible content without asking permission and without paying a license fee.
The principle behind it is simple. The right to read must include the right to mine. That's the basic idea. Facts and patterns are not property. A researcher analyzing 10,000 papers to find a new drug target with AI is not competing with the authors, she is advancing the very purpose for which those authors published.
While this freedom to mine is clear in principle, its implementation is muddled in practice, unfortunately. Rights holders can reserve the rights for commercial mining, and uncertainty persists about how reservations must be expressed, where research ends and commercial activity begins, and how the exception applies to public-private partnerships, which are now common Europe science models. Technical protection measures can quietly hollow out in practice what the law has granted in theory. And a researcher who fears a lawsuit will simply not run the experiment. So we must shore up this protection.
Firstly, we should clarify that contractual clauses and technical measures cannot override the research exemption. Secondly, we need to give legal certainty for public-private partnerships, whatever that may look like in practice. Thirdly, we should ensure that the exception is genuinely usable by independent researchers in small institutions, not only by those with legal departments. Open science cannot mean open only to those with an army of lawyers.
So, this would be my pitch in the beginning or my ideas and where we try to work on with the initiative report, but for sure will also work on the legal framework once the Commission comes up with it. The two key challenges I've addressed now are different, but are united by a common idea. Intellectual property law should always stimulate human creativity and serve the public good. And this is the balance we have to strike. It should never become a wall to stifle them.
The knowledge commons needs to remain an open frontier to everyone. And I think it is our task here in the European Parliament, but also wherever this issue comes up, that we have to ensure that all Europeans can continue to advance it. This is the foundation of our success and I think we have to protect it. Thank you very much for listening to me and I'm very much looking forward to the discussion.
I will not be able to listen to all of that because of other meetings, but thanks for giving me the opportunities to speak here and I wish you a very good conference here.
Speaker 8
So now as a next speaker we have Professor Richard Gold. Richard, can you hear me?
Richard Gold
Thank you. Thank you everyone for inviting me today all the way from Canada, although I do have European citizenship as well through France. So I'm speaking to you both from afar and being with you. As Katerina mentioned, I want to talk about two things in particular, and that is the cohabitation of intellectual property and open science, how properly structured they actually work together, and then I want to talk about the importance of open science as a way to ensure our mutual sovereignty and working together we're actually more sovereign.
The basic point on intellectual property and open science is they're both aiming at providing financing and innovation, but through different means. Intellectual property raises the returns to those who are inventing, while open science lowers the cost and risk of inventing. So these are parallel processes. And we need the two of them in balance. And frankly, over the last 20 to 30 years, the balance has been much too heavy on intellectual property and not a lot on open science, but that is starting to reverse.
So what do we know about research? Well, one of the issues, and most of this data is American, I apologize, that's where we have much of the data. What we're seeing is that research productivity is declining. That is, in an important 2020 paper, we've seen that you need 41 times more researchers to produce the same output. And it's not that we're getting fewer patents, but those patents are less accessible.
That is, there are a few large firms that control them and they're not allowing SMEs to really participate. And that is having a negative impact on the flow of ideas and our ability to build innovation. And so our central problem is how do we turn ideas not just into patents, because patents are just a tool, but into growth. And we do that by sharing and testing and working together. So AI, as just mentioned, is an important tool that's coming on, but we have to think about how we structure it.
And so I appreciate the previous speaker's presentation. That is, if we're not only thinking about AI development, but the underlying scientific enterprise and how we organize science, particularly using open science, we will never gain the advantages that we hope for. Okay, let me return to that first slide about intellectual property and open science. So what intellectual property does is it provides exclusivity over certain inventions.
And the idea is, knowing that I'm going to be able to obtain this exclusivity later on, I'm willing to invest today. So it allows inventors, it allows companies to invest today in the hopes of recouping that cost of research. It creates tradable assets so that if I'm a smaller company, I know I can sell that asset to a larger one later on and therefore recoup my costs. Patents do require some level of disclosure, but patent agents and companies have become very adept at limiting the value of that disclosure so that people can't fully build on it without doing significant complementary research. On the other hand, open science lowers the costs and the risks.
And it does so by pooling early-stage risks, so rather than me having to absorb all the risk of doing the research, I can share it with others. So one euro of investment gives me ten euros of pooled investment, and therefore I'm willing to take on higher risk because my cost is lower. We also develop shared tools and data that decrease duplication so that company A doesn't do research, find out what they're researching doesn't work, and then company B try to do the same thing because they didn't have access to it, and then company C.
There's a lot of duplication. By sharing these tools and data, we actually cut duplication, and that means we cut costs. It also simplifies contracting relationships because we know that intellectual property, particularly with universities and research institutes, can delay project start, let alone completion, by six months to a year or more. So what we have seen, and the evidence is still sparse, I don't want to overplay this, but in experiments, natural experiments, and people trying to create open science partnerships, open science partnerships are public-private partnerships built around the principles of open science, we have seen some benefit.
So there was a Heidi Williams is a renowned economist in the United States and she studied the privatization of genetic information by Celera, comparing it to the open human genome project. And she found that there was a 20 to 30 percent decrease in follow-on research where the knowledge was kept in-house, protected at Celera compared to the open approach. So we can see here that people stay away from research areas if they're already encumbered by intellectual property. We know that public funding of research provides much more both public sector but also private sector development. We see 2.3 additional private sector patents for every million dollars of public biomedical research.
And we've seen in those few open science partnerships that have led to the production of drugs or other products that we can save about two years or up to two years in development time by open sharing, by decreasing the negotiations over intellectual property. We know that early stage cooperation raises R and D investment and increases the odds that a project advances as we wish.
IP and open science also have an effect on who benefits from innovation. So IP makes private companies able to invest and recoup very costly research, particularly biomedical, but it comes at the cost of higher prices during the exclusivity period, so 20 years plus up to five other years. It leads to evergreening, which is forever patents. It leads to patent thickets, which are difficult to negotiate through. Secrecy creates data silos and has a negative effect on follow-on research by other firms.
Open science, on the other hand, works by making tools and knowledge free to use or almost free to use, lowers the entry costs, particularly for SMEs, academics, and the global south, and makes that knowledge accessible for further development. It doesn't guarantee, however, that downstream research will occur or that it will be affordable. Other programs are needed there.
And where we see openness as having the most direct impact are in the biomedical field, rare and neglected diseases where we already see that firms are pulling out, complex diseases such as neurodegenerative diseases where firms are pulling out, and frankly AI. If you're not open in the AI field, you're not really playing the game. We also see, and this is US data, European data is a little bit different here, is if we look at the patenting that our firms, and I say our by meaning both Europe, Canada, and a few others, in the United States, we contribute collectively to a lot of that innovation and those patents, but we don't own them. So in the case of Canada, we own about 50%, Spain is closer to a third. That means that for every three patents issued in the United States with a Spanish inventor, only a third of those are owned by a Spanish firm.
And that means we are making, we are investing in research that leads to patents, but a lot of those patents are controlled by US entities. Open science also has a positive effect on sovereignty, research sovereignty and national sovereignty.
One of the biggest risks in the world we see today is being cut off from research. We saw that when United States decided to close down its climate data, suddenly researchers were scrambling to find it. We saw it during COVID when certain, United States again, was not willing to export vaccines, so Canada actually imported from Belgium, not from United States, even though we're much closer to United States. And so the danger of being locked out, especially during a crisis, is, could be extremely detrimental. If we don't have availability of data and tools during a crisis, it means that our firms and our researchers can never develop the response that we need.
And so open science ensures that this data is open. We need the infrastructure to house that data and make sure that we're never physically locked out. We can have sites in different countries to ensure that. But it is incredibly important, especially for middle powers, to be able to ensure that they'll never be locked out and that their researchers will be able to go forward.
We also see that whoever sets the standards of data and tools sets the playing field and has an advantage. And Europe has many such platforms that set standards: Elixir, OpenPhacts, IMI, and others build infrastructure and data that sets global standards. And those standards mean that your firms, or our firms, are able to access that information less expensively and faster, which means that we have a competitive advantage. At the same time, while open data is there and open tools, people carry much of the knowledge needed to innovate. And that tacit knowledge stays in people's heads, and so they stay local.
So if we have attractive cities, as there are many in Europe, then people want to stay there and they build a hub. So you can have openness but still retain talent locally that allows your firms to develop products and services at a global scale.
And cooperation between, you know, Spain and Canada, Denmark and Canada, Germany and Denmark and so on allows us to each focus on our areas of strength and pool that knowledge so that we're all better off. So what we need is not to abolish IP or open science but to make them work better. We need to rebalance. And to do that, governments need to fund the commons. We need to fund these public-private partnerships that aim at developing certain products.
There's the Structural Genomics Consortium that is in Europe, UK, Canada and elsewhere developing knowledge around the three-dimensional structure of proteins and other things. We need open data platforms, but our funding tends to focus on either individual small research products or on getting patents. And we need to think more creatively about opening our research funding to specifically fund these partnerships. We need to ensure that upstream knowledge doesn't get patented too early, doesn't get absorbed by large multinationals, a large number of which are based in the United States, that curtail the flow of knowledge needed to support our SMEs. We need to find structures within universities and research institutes to reward sharing.
Currently we don't do this. We count publications, some institutions count patents, but we don't actually reward people for developing common standards for data, publishing negative results, sharing the knowledge generally. This is not a free exercise, it takes time and we do not provide the support necessary, nor do people advance within these institutions based on it. We need to change those metrics. And we have to measure, therefore, what matters.
We need to look at those rewards, but we also need to provide the research community with the data to support the right balance. We need data on trials, on patents, on licenses, on public-private partnership that is available to government but not necessarily shared. So the European Patent Office could certainly provide better data on who's patenting and licensing, they do some.
The funding of research could be easier to find in terms of who's working together and so on. So bottom line, intellectual property rewards the inventor and this is critically important to get them to invest. But open science is also critical because it lowers the cost of inventing. And often what we will see then is earlier stage pre-competitive and that the boundary between pre-competitive and competitive is moving. That could be open.
And then we allow firms to compete afterwards. We can cooperate between countries in those earlier phases and then let our companies go off and use that knowledge. Europe has already heavily invested in open science through the Open Science Cloud, but actually the partnerships part is missing. And given many of the strength of Europe in terms of these platforms, data platforms and so on, you have a strong base for moving forward. So I will end there.
Thank you very much.
Speaker 12
Yeah, thank you so much Richard, back to you in Canada. We take it back to Brussels and our next speaker will be Gustaf Nilssonne.
Gustaf Nilssonne
Thank you Katarina. Good afternoon everyone.
Thanks for the opportunity to speak here today. So, I'm Gustav Nilssonne, I'm an associate professor at the Karolinska Institute in Stockholm, Sweden, and I'm going to talk today about our Horizon Europe-funded project IP4OS, unpacking the possibilities of intellectual property for open science, in which I've had the great pleasure to lead one of the work packages and to work together with many colleagues, including Katarina and many others who are present here today. So, I wanted to start by giving a few concrete examples of the kind of questions that we've been looking at. I have three scenarios here of researchers who are facing the intersection between intellectual property rights and open science. And first we have Alice.
Alice has written a manuscript and sent it to a scientific publisher, and the publisher asks Alice to sign away her copyright to the publisher. And Alice now wants to make sure that other researchers can read the paper, while at the same time she needs to advance her own career, and she, in order to do that, she has to publish papers. Bob has made a research discovery.
He has created a new material that has some interesting characteristics and he wants to disseminate these findings as soon as possible. But at the same time he wants to protect his possibility to patent the material and to valorize it together with an industrial partner. Charlie has collected research data. Charlie wants to share these data according to the FAIR principles and the principle of as open as possible, as closed as necessary. But she is concerned that if she shares the data completely openly, freely, she might give up some intellectual property ownership or academic recognition if someone else then uses the data to publish.
So here we have three case examples and they relate to different aspects of intellectual property rights as you can see. One of the examples is about copyright, one is about patents and one is about database rights. And these are some of the main clusters that we have been looking at.
So, I will now very briefly try to explain what it is we have done in this two-year project, how we have conducted the research and mainly what our outputs are. So, we have investigated how intellectual property rights management and open science can work together. And we have developed what we call a synergy framework, which is a document with recommendations. And this is of course in service of the aim of knowledge valorisation. And this research, if I may say so myself, is very timely in light of the development of AI, which has of course become extremely prominent in policy debates in the last years.
And also, I would argue that it informs the current debate on open science in relation to research security, which is another theme that has gained a lot of traction during these two years that our project has been going on and which I will come back to. So, briefly now, our research approach has three strands. We have conducted a scoping review of the existing literature on intellectual property rights management and open science. We have run a survey of stakeholders in Europe and we have brought in expert opinions. So, our scoping review is a very extensive body of research where we have included 266 papers.
This work was led by Gautam Sharma, a postdoc who is here in the audience. And very briefly, you can read the whole thing, but some quick takeaways are that there is a set of challenges around unclear ownership, restrictive licenses, and differing legal rules as perceived in this body of work. Researchers face concerns about losing credit, losing control, and losing publication opportunities when sharing data and other research outputs. But there are also solutions. And these include open licenses, rights retention strategies for scientific manuscripts, controlled access, mainly for sensitive data, and selective protection.
And these are some of the conceptual themes, let's say, around solutions that have emerged in this review. Our survey included 176 respondents, stakeholders, researchers, research managers, and so forth in Europe.
And very briefly here we find that among our respondents, the synergistic use of IP and open science is not very common, although respondents think that it is indeed possible. There is a perceived lack of policy tools to bring together intellectual property rights management and open science, and there is also a problem concerning limited awareness and an uncertainty about the rules. And this is perceived by some as a barrier in itself, so that researchers, for instance, when they don't really know what to do, may choose to be more conservative and put the data in the file drawer instead of publishing the data. Or in the case of patents, the opposite may happen: researchers may publish too soon and may lose the possibility to protect their ideas with a patent. This also you can look up of course in detail if you wish.
For expert opinions, we have of course our own internal research and we have also had a large number of webinars with international experts and testimonials, video-recorded statements.
Now I will give a lightning overview of the recommendations in our synergy framework. So here we have collected best practice examples in the fields of copyright, patents and database rights and we make a number of recommendations directed at different stakeholder groups. So for researchers, we recommend first of all to decide to be open as a point of principle, as a starting point. We recommend that researchers learn about copyright and licenses. That researchers assess in their projects early on whether they have ideas or innovations that could be the subject of patents so that patents can be applied for at the right time and to sequence this in the correct way and apply for patents first and then publish.
And also in collaborations to define the rights, the rights to data ownership and access rights and reuse conditions. For policy makers, we recommend to establish policies for intellectual property rights and open science which go together, to consider them in relation to each other and preferably as one whole rather than two distinct things. We recommend support to researchers through multi-professional teams, which is something that we have ourselves been building up in this project, and to develop infrastructure using, for example, patent pools, open innovation, cross-licensing and clearing houses. I know that I'm going very fast, but again, all of this is published and you are very welcome to look up the details. An extra recommendation to policy makers and institutions, and this ties in with something that Richard Gold mentioned in his presentation a few minutes ago, to reform research assessment to better value the open sharing of research outputs.
So, for example, many researchers perceive that we are now being assessed by the number of manuscripts that we publish and the venues that they are published in, and this cannot but shape our behaviors as researchers. We have to try then to publish manuscripts in that way in order to carry on and to further our careers, whereas, for example, sharing research data is not generally perceived as much as something that is valued.
And this locks us in in the current publication system and it limits our opportunities to prioritize open sharing of research outputs. And thirdly, for policymakers, to define unified and transparent national policies on research data that go together with EU policies. For capacity building, we recommend training on intellectual property rights and open science for all stakeholders. We recommend training for researchers in copyright, privacy, and trade secret considerations, guidance on licenses, and especially the Creative Commons licenses, which are widely recommended for sharing research outputs. And to equip multi-professional teams with a variety of professional roles to support researchers.
So that's a whistle-stop tour of the main recommendations in our synergy framework. And we have delivered this synergy framework along with a number of other outputs. We have a toolbox for research knowledge valorization with a number of components that can be used by researchers, by research support professionals, by leadership in research institutions. And the toolbox, together with the synergy framework and with all our outputs, have been openly published on Zenodo and you can find them there.
And the toolbox is also available in the EU's knowledge valorization platform. And we hope of course that these resources will be a help to researchers and to others going forward. Lastly, I want to say something briefly about future perspectives. So we have not solved all problems with this project and there are other challenges that are emerging, not least related to AI. So some points that I wanted to put on the table and that we might discuss.
AI is changing the way that research is being done and it's possible to generate a lot of things. But AI also needs data as inputs, as Timo Völken emphasized. And I think that we need to find new and better ways to follow the life course of data so that we can verify how data were generated, when and where, where the data went next, how they were curated, at what point they were ingested into a model and so forth to be able to ensure that the data are real and true in some sense.
We also need to improve the capabilities for researchers in Europe to use AI. And for this, we need to invest in infrastructures, strategic public interest AI for science infrastructures that are rights-aware, so that as a researcher I can know which of the data that I can use, which of the data I can feed into a model, and which of the data that I have generated that are made available to others. We suggest to support an EU-based interoperable system of systems, and this of course is more or less what the EOSC is trying to be, a system of systems to connect data, publications, a variety of research tools and outputs. We suggest to align policies on open science, on AI governance, on research integrity, on copyright, research assessment and infrastructure. All of these things go together, and it's about making it possible for researchers to do our best work and to do so openly, but also to make sure that we are rewarded for it in a way that is aligned with what we collectively are trying to achieve.
And lastly, a small point on the relationship between open science and research security.
So research security can mean different things. It can mean that we don't have other countries stealing our outputs, but it can also mean that we make sure that our data are actually not tampered with, that we can guarantee the integrity of our various data and research objects. And the best way to do that in most cases is to publish the data openly in a trustworthy repository. And so I want to also highlight that it is indeed possible and in many cases very much the case that open science can support research security rather than being in tension with it. Thank you very much.
Speaker 14
Thank you so much, Gustav. And now we come to our next speaker, Professor Puigdomènech. Please, the floor is yours.
Pere Puigdomènech
Thank you very much.
I'm going to present what I think are the problems surrounding one important aspect of intellectual property protection, which are plants that at this moment can be modified by new genomic techniques. One of the points that we have is that... Present European legislation is this one, is Directive 98/44. So it was published in 1998. And you see from 1998 many things have happened in the field of genomics, in the field of molecular biology, many things.
And the directive is the same one. And one of the reasons is that it took a lot of time to have a compromise for this directive, which is a very complex one.
This we have to be, especially in this house, to be well aware of the complexity of these discussions. And one of the aspects of this directive is that the following shall not be patentable. And you see plant and animal varieties cannot be patented. Or essentially biological processes for the production of plant and animals. And inventions which concern plants shall be patentable if the technical feasibility is not confined to a specific plant, you see, a plant variety.
So here there is a compromise, a complex one, but it seems very clear that plant varieties are not patentable. Why it's so, in part because there is another system of intellectual property protection in the world, which has been, you see, working since 1961. In fact, intellectual property of biological systems have started in plants. Why has been started in plants? Because you see to have a new plant variety is important for two special features.
Pere Puigdomènech
Yield production has increased enormously. A new hybrid from maize may allow the farmer to have eight times more production for the same hectare than a non-modified, a non-bred one. So, there are a lot of new pests arriving, new diseases arriving. So, we need very, a lot of inventions to have our food production to keep pace in this situation, number one. Number two, it takes a lot of time to have a new plant variety.
It may take 20 years to have a new plant variety, you see. So, you have to reward, it will be very complicated if someone spends 20 years to have a new plant variety and someone arrives and takes this variety and has the benefit of it, okay? So, that's the reason why so early as 1961, it was this international convention.
This international convention, the most important thing is that it defines what can be protected by this convention. It's a variety that has novelty, something new, distinctness, it's different from those existing, it is uniform, so it is useful for the farmer because it's uniform and it's stable, so it can be used different times. It has a lot of different aspects. Breeders, for instance, that is very important, is an example of open science because breeders can use any variety that exists, even if it's protected, but if it changes enough to have this novelty, distinctness, or these properties, they can apply for the new variety. So it has, it is, it has these specific features, this convention, and is one of the reasons probably why the directive of patents avoids to take this kind of protection because there is another one which is working.
So, as this is relatively complex because people started to try to patent different systems, different, for instance, plants.
There was a couple of patents on plants, not plant varieties, but on plants, for instance, to plant broccoli, for instance, or something like that. So it was a report that it was published in 2016 that it has these questions, no? These questions on plant breeding, breeder's exception. Breeder's exception is that, as I told you, any breeder can take a variety to have a new type. But this final report, it was relatively complex and one of the examples is that recommend no action is taken.
You see, because it was thought that maybe to touch the directive of 1998 could be even worse, you see, at this moment. But you see, science is going very fast. The cost of DNA sequencing, you see, you see this is a logarithmic scale. When we work for the first human genome in 2001 or the first plant genome in 2001, the European Union paid scientists one euro per base pair.
You see, that was three billion euros for the human genome. Now you have it for 1,000 euros. So the cost has been reduced by one million, you see, in the last 25 years. It's not bad. So things have changed enormously.
And at the same time, this information, which is open, the rule to publish any DNA sequence is to have it open in an open repository. We have a repository here in Europe, another in the States, another in Japan, and it's completely open. Of course, it has the, it opens the question whether something that is public and open is being used by AI systematically. That's one of the questions that's open. But number one, cost of DNA sequencing has decreased enormously.
Second, there are new methods of changing genomes.
And now there are AI systems that use both things. But you see, it was in 2012, the publications. But more or less at the same time, it was a dispute for patents. Because, well, the technicalities are relatively complex, but there has been from the beginning a dispute between Berkeley and Harvard. Berkeley plus Vienna and Harvard is the Broad Institute.
And the patenting for the genome editing has started from the beginning. And this year, you see, it has, you see, it's March 2026. The Broad Institute says it is clear that the patent for genome editing is for the Broad Institute. They are holding the patent, not the Vienna, not the Berkeley. In Europe things are a bit more complicated, but this is the result of the court decision this year.
And if you see, they say that they are going to license for plants. From the beginning, the Broad Institute has licensed the use of genome editing to a big company, American company, Corteva. And you see, it says Corteva, they are open innovation, but making the CRISPR-Cas9 technique available.
It is not clear in what conditions. And I put here a picture that highlights that Corteva has split just this month. And the business for seeds now is not part of Corteva. They change fertilizers and pesticides one side and seeds another side. And we don't know in what conditions the license is going to apply and who is going to decide about this license on plants from CRISPR-Cas9 that has been licensed by Broad to Corteva.
So you see that these things in plant varieties are changing and complex. But now it seems that at least from the American courts, the license for genome editing of plants is in the hands of Corteva or Bayer, we will see. So I am a member of ALLEA. ALLEA is the Association of European Academies.
And we were asked by the Swiss Academy to try to analyze the situation of the IP system on new genomic techniques for crops. Why? Because farmers are worried. Seed companies are worried. There are in Europe, we have a system of seed companies which is very complex.
They are small companies that try to produce seeds for specific countries or for specific users. There are big companies that have a lot of seeds from different points, different crops. So, what ALLEA did was to try to identify the problem. Why it seems that there is a problem? And it's that what they say here.
Breeders may not have access to all the genetic material because it has been protected differently from what happens with the UPOV Convention, you see. Second, there is traditional breeders may have to defend themselves for using something that appeared in the field, but it's also at the same time produced by new genomic techniques. You see, we have the two things together and people may have to defend.
And small breeders are worried. You see, it's expensive, these kind of things. Maybe not to have the required license because the conditions for granting a license, they are not clear, or unintentional patent infringement if it is not exactly, you may have, breeders have this custom of using everything which is in the field and they try to make it novel and different. So they may have unintentional patent infringement. Those are the reasons why we, so what we did, we talked to all people, to farmers, to companies, to big, to small, to lawyers, to scientists, and so we try to make a number of recommendations at the short time trying to facilitate access.
At this moment, it's very difficult to have exactly how and where you see that this big company now is changing, so we don't know exactly the conditions. Mandating patent database, it is also there are associations of seed companies that have databases for patents, but not all companies are using them. Stricter interpretation of patentability, you see that there are disputes on this, what is patentable or not. And structured licensing schemes for publicly funded research and ethical licensing. You see from time to time some of these varieties are very important for production, you see.
So from an ethical point of view, it could be important to try to allow this kind of use of such a broad type of technique. So we thought that those could be short-term measures. Mid-term, suspending patents, like there are people who say we are talking about food, so we have no right to put a barrier for use in some cases, no? So let's suspend some patents for some types of uses.
Companies have already started to do that, to make patent pools, trying to exchange, not to make barriers for using that. And standard essential patents, to identify what are the essential ones and try to use these ones. A long-term measure will be compulsory licensing, which is in the law. It's complex to use it, it's complex to define. Amending the European Biotechnology Directive.
This is the mantra, something that in this house they will think twice before doing that. But probably at some point it will be necessary to start these complex things. And finally, a new type of IPR for plants, as it was already in the compulsory agreement. So that was me, and you have seen that I have not talked at all about regulations. In this house, they have been approved this year a new regulation for new genomic techniques of plants.
And I have not talked because I think they are two different questions. One question is regulation for health and safety, another question is the patenting of plants. The questions, I think those are different, and they don't have to be mixed.
Thank you very much.
Speaker 19
Yeah, thank you so much, Professor Putch-Domenec. I think that was very interesting, very important, very technical, but we see many changes are needed. And we come to our next speaker, who will now introduce us to the importance of the capacity building, Annika Müller de Carabil, please.
Annika Bruna
Thank you. Good afternoon everyone, thank you Katharina for your nice welcome. Thanks everyone for being here, it's quite a crowded room, I like this. And I see so many young faces also, that motivates me a lot. Also, I like all the other faces, I didn't mean to say that.
Speaker 17
I will be speaking about open science capacity building and IP through the lens of the findings of the report of the Open Science Learning Gate 2025. That report has just recently been launched. You all very welcome to look at it, we're happy to share the QR code and all the information at the end of the presentation.
So I will start with the gate, with the Open Science Learning Gate. What is it? What's the report about? But then we will directly jump into key outcomes from the report for capacity building and IP, and I will close with how the gate can support processes. And what's interesting, because intellectual properties are obviously our big topic.
What we see in our findings, just to give you a very brief glimpse into it, is that it plays a small role and at the same time a growing role. And then we see a signal that has been mentioned a lot today already, that since AI entered the stage, the awareness seems to be growing for intellectual property, may it be licenses, may it be patents, may it be anything that you want to put your stamp on it. So what's the Open Science Learning Gate? It's a capacity building infrastructure, and our tantrum really is dialogue.
So it's based on dialogue with experts and with practicing communities, bottom-up approach. And the main aim is to screen the current state of open science and practices and to see how it evolves. The report is built on the evidence that we gathered, and it's all openly shared on Zenodo as a static version and on Data Studio as a live version. This is what it looks like when you go to the Data Studio. You don't have to read this now.
The only thing I want to tell you is that we summarized our main findings in six statements, and we underline all the statements with the accompanying evidence, so you can read through when you find the time. I will not go through the six statements now, and I will not go through them due to two reasons. Because on the one hand, I was told to speak for seven minutes, and it's six statements, so that's difficult. And on the other hand, the report is on open science in general, so not everything focuses around IP and AI and capacity building. But there are strong signals, so what we did beforehand is we looked through all the statements, where do they directly or indirectly touch upon the topics we're interested in today: IP, AI, capacity building.
And we also looked how are all these dots, all these signals that we see there, connected to each other to give you a brief overview what the findings signal towards intellectual property and IP and capacity building. And that's a bit, well you can still read it, I think it doesn't matter. But what we see is when we summarize, when we synthesize the findings, we see a gap. And we see a gap between concepts and instruments or tools. And that's not a new gap, that gap has been addressed before, but we're reconfirming this gap basically.
And what we see is frequently discussed in recent open science resources, that's concepts. And with concepts we mean fair data, open research, transparency, as open as possible, as closed as necessary. But what on the other hand is less addressed in current open science resources is instruments or tools.
And that goes for data protection and that also goes for IP protection. Now what's interesting with IP is that it's less represented in general open science materials, but it seems to be more represented in open science resources that kind of dance at the interplay of open science and artificial intelligence. So again there we have this signal to see that, ah okay, AI has entered the stage and now we clearly looking into the topic more with depth, with more interest, huh? So here we see a how-gap so to speak. And at the same time practice doesn't sleep.
So in the solution sessions, in the workshop that we do for the gate, we see that the communities already come up with a lot of ideas on how to work around this from their perspective missing literacy on instruments and find solutions for safe data sharing in repositories, data management and so on and so forth, while at the same time, and then we come back to capacity building, they also say there is a call, we need more education and guidance on instruments.
So capacity building needs to put a focus, and it does already, so we've heard great examples today already. I'm just saying the focus still can be increased. There's room to the ceiling, how you would say in German, I don't know if that translates into English, but I think you understand what I mean. Licensing was an example here, training on instruments meaning licenses or also data management plans. And there's a lot of training in place as I said, but that's not the end of the road.
So what does this mean for capacity building in the IP framework? What we see really when we summarize is a kind of bottleneck. So we see there's a lot of knowledge when it comes to concept, that's broad, it seems to be broad, but we seem to be approaching a much narrower zone when it comes to putting concepts into practice. So to give you a very practical example, I might know about FAIR data and I know how to make my data set FAIR. But at the same time, when I come to the point to make the decisions, okay, what about reusing my data?
What about licenses? What about at what point of my data management plan do I need to actually think about all of this?
And that is something that applies to capacity building. So we see a need for improved IP literacy, but also a sustainable dialogue that includes IP in the open science dialogue. And that's where the GATE comes in, because the GATE learns from practice. So practitioners, experts, they share their knowledge, they share their materials, they give feedback. The GATE takes this knowledge, feeds it back to the community after analyzing it, but also translates these insights into capacity building structures and ways.
It can connect the expertise needed to act. So we heard a lot this morning about multi-professional teams and how all this expertise needs to sit at one table in order to give good counseling. So there needs to be a lot of expertise.
And most important, I think, is what the gate can really support the structures with is keeping the dialogue and the learning alive beyond projects. So we've heard about the great IP4S project, but it's a project and it ends and it has great outcomes. But what do we do with the outcomes now? So the gate can be seen as a long-term capacity building infrastructure. And it is, by the way, a sustainability partner of IP4S.
So it will try to keep alive what's been done from great work there. These are my takeaways. So we've seen a gap. Concepts are well addressed. Instruments, also intellectual property protection tools, licensing are less addressed.
There's a call for more informed capacity building that includes IP literacy. And AI seems to make this more urgent. And the gate can play a sustainable role in shedding light on practices, shedding light on gaps, needs, and support targeted capacity building with continuous evidence. If that sounds interesting, we're happy for you to read our report, reach out to us, we're open to collaboration. Thanks a lot for your attention.
Speaker 19
Thank you so much, Annika. And now, the missing piece, the last perspective, yeah, is now coming from the European Commission. Thank you so much, so the floor is yours, Manuel, wherever you want, you can sit there. Yeah, yeah. Thank you so much.
Manuel Aleixo
Thank you very much, good afternoon everyone, and first of all, thank you very much to the European Parliament, to Lina Gálvez, and of course to Catarina and her team for putting all this event together. And thank you to all the speakers because I found it very, very interesting to listen to your insights, to your thoughts, to your perspectives, and I'll explain in a moment why this is important and why this is a good moment for us. Let me begin by saying that I'm a little bit disappointed with Gustav Nilsson's presentation because I was waiting to hear what the outcome of Alice, Bob, and Charlie's questions were. And so you left us hanging a little bit there, but I think we can figure it out as we go along.
So let me introduce here a slightly additional perspective to this topic, to open science and to IP. In the Commission, we look at open science from the particular point of view of publicly funded research. And indeed, what is the situation today in EU R&I policy and funding? So under Horizon Europe, you will probably know that we have provisions in our program that require that all applicants that receive funding from Horizon Europe make their publications available in open access and make their data available, of course, with this caveat as open as possible and as closed as necessary. So these are the rules under Horizon Europe.
The data tells us that we are at around 82% open science of research outputs funded by Horizon Europe. So we're doing well. We could be doing a little bit better. And of course, we want to go for 100%, of course.
Now this is the situation in as regards publicly funded research under Horizon Europe, but what about the member states? And this is also an interesting thing to look at because it's estimated that across Europe we are at about 65% of open science in terms of publicly funded research at the level of EU member states. So 65% is not bad, but of course it's quite far from 100%. And the logic, the public policy logic behind open science as far as the Commission is concerned is that we believe that because science is publicly funded, the taxpayer should not have to pay twice to access the results of that research. So if you like, there's definitely a public policy objective behind this.
We believe that it's fair that when taxpayers pay for the research, that it's fair that they can access the results of that research.
Now, naturally, under our rules, under Horizon Europe, we have provisions that allow for certain exceptions, and of course that includes IP protection, and that includes, for example, restricting the disclosure of research outputs that may have other implications, for example, of security implications. But by and large, we strive for having research outputs funded by Horizon Europe publicly available. Now, in the member states, the situation is a bit more complicated than that. I mentioned that we were at about 65%, and part of the difficulty is that in some member states you have legal provisions on open access obligations, in other member states you don't have at all, and in those who have legal provisions, they tend to be quite different from each other. And some member states, two of them, actually rolled back their open access requirements in recent times.
So we have an uneven situation across Europe in what regards open science, not science funded by Horizon Europe, of course, but science funded at the national level, which as you know is most of the science in Europe.
Horizon Europe represents about 10% of all the public publicly funded research in Europe, so it is a small part. It's a big program, but it's still a small part of public publicly funded research. So all of what was said here about the need to protect IP, the need to take into consideration research security issues, for example, which is a very, very good point, and I can say that a bit more about that in a second. The need to take into consideration research assessment, which is also connected to open access. And of course, the efforts that we make in developing EOSC, which is the European Open Science Cloud that has been mentioned here, which is about federating structures across across Europe to allow precisely repositories to exist throughout the member states.
So just a couple of words about research security and about research assessment. So on research security, we've developed in recent years, we made a proposal for a Council recommendation on research security, and this was adopted by the member states. This is not legally binding, but it has a number of recommendations, recommendations to the member states, but also recommendations to the Commission. So in a way, we proposed to recommend ourselves to do certain things. And and indeed we are doing these things.
We are putting in place a center of expertise on research security and a due diligence platform, among other instruments. And the purpose of these instruments is not to provide legally binding rules to the Member States or to research performing organizations, universities, etc., but rather to provide them the means with which to take decisions. Should they collaborate with this or that entity? Should they collaborate with this or that university? In order to make those decisions, they have to have, of course, the information.
So this is our perspective. Rather than interfering in national legal regimes, we prefer to provide the information for entities and institutions to make informed choices. So we don't foresee any particular legislative intervention on research security at this stage. The same goes for research management. On research management, we have, sorry, research assessment, we have an initiative called CoARA that you've probably heard about.
And this is about, you know, a bottom-up voluntary initiative bringing together hundreds of universities, research performing organizations, etc., in order to examine the issue of research assessment and how this can change and how that change can be then enacted in their institutions.
So again, on that, we don't foresee any particular top-down hard and fast legally binding rules to be applied to the Member States and to institutions and ultimately researchers. We do have, however, an initiative that we are planning and that will be proposed towards the end of the year, hopefully, which is called the European Research Area Act. And this is about trying to develop the European Research Area, what Enrico Letta calls, not originally, the fifth freedom. And so this is supposed to be legislation, legally binding rules in order to allow the free flow of knowledge within the European Union. Now, as I say, this is still under preparation, so I can't tell you for sure what you will find in the European Research Area Act, but we are working in several, looking at several areas about investments at the level of the Member States, about coordinating better investments among Member States because we have to address fragmentation and create critical mass, about the free movement of researchers within Europe, but also about the free movement of knowledge.
And this also implies the freedom of scientific research and also eventually open science. So we are looking at the possibility whether it does or doesn't make sense to have a legislative intervention, to have legally binding rules at EU level that apply to all the Member States to create a level playing field on open science.
I'm not saying we'll do that necessarily but we are looking at that. So this discussion is quite interesting because it puts open science into perspective and in this case looking at open science in perspective with of course intellectual property rights protection. But as I say from the Commission's point of view our perspective is that open science that is an imperative because publicly funded research is funded by the taxpayer and should be publicly available for all the reasons then that we heard about the importance of the outputs of research being publicly available for science to develop. Maybe one last point I think the the work that IP4OS is carrying out I think will be useful to feed into the work that we are doing at the Commission and of course the the information that you're sharing so thank you very much for that.
We do need this evidence, we do need to understand what is it that researchers need ultimately and what justifies or doesn't justify an intervention at EU level, at the legislative level, but also it can be at other levels. It can be with soft law or it can be simply with capacity building measures, with funding funding projects, etc, etc. So that is quite important and and this dialogue should definitely continue and and and for the Commission it's extremely important. So you can put in your calendars towards the end of the year, hopefully, we will adopt the proposal for the European Research Area Act and then we'll see whether we have something on open science or not. Thank you very much.
Speaker 19
So thank you very much for this missing piece. I think it was crystal clear, so we're looking forward to this. I have it in my calendar as of today. Thank you very much Manuel. And now we open the floor for questions.
And I just wanted to set the scene to lower your expectations. We might be able to take three questions, it depends on how long they are or the replies. So who would like to be the icebreaker?
Angelos Karageorgos
Thank you, Mr. Chairman. Good evening, citizens. Angelos Karageorgos, Ecopolis advisors. We represent the Ecopolis movement, a pan-European and Mediterranean movement based on the Aristotle's principles, a meta-communistic and post-capitalist one, which is based in direct democracy, citizens' assemblies, science and societies of knowledge, and arts and ecology. Of course, we thank you for the interesting theme here, but I think you didn't respond about rethinking intellectual property in the era of open science.
We didn't get anything about that. I think in the way you presented the issue, and excuse me for that, it's outdated if we compare with the other side of the Atlantic Ocean. Because first of all, open science to work very well in a globalized environment, which today is not functional and is going to follow maybe the directions we give through the ecumenism, which is a different model of the globalism. And open science can work actually in closed ecosystems. In the global thing, it's not workable because there are different political systems which they are trying to get the information for their benefit.
So it's not workable. The second thing which we have to to observe, we observed is the the limits between the open science and the AI are not so well described because the AI, the generative AI, the training models can use a lot of data which they are under copyright laws or different things of different ownership of data. And this cannot be proven that a human being was interfered so that the IP to be functional according to the legislations. So you didn't respond at all about, you didn't speak at all about that. And also the the other side of the Atlantic is is the discussions right now are who owns the data so that the the generative AI to use them has the ability and the legislative functionality to use them.
Ending our intervention is let's see the open science how it worked in the Linux system with the Linux Foundation, for example. When we got a European open science data language and operational systems and some companies they got it and they use it on their way so that nowadays the open science, which is for the broader scientific community, is used by only some companies. Thank you very much.
Speaker 19
I am thank you so much. Yeah, I always encourage my students to be very disruptive and to be have to always have good questions. So you always talked about the other side of the continent. So Richard Gold, you are at the other side of the continent. Would you feel would you like to take this?
Richard Gold
Sure. I mean, open, you know, we only had two hours. We can't address everything. I think if I could just deal with a few of the things. I mean, AI is certainly causing us to rethink issues around intellectual property.
One of the issues is that creators, artists, writers, and others are very concerned about the use of their writings and artwork in AI, and countries are struggling to figure out what's the right balance. At the same time, we want AI companies to be able to train. And so that's an ongoing discussion. Do we create exceptions?
Do we use what we call fair dealing in Canada, fair use in the United States, to allow it? But more importantly, the next stages of AI, I think, are going to be in science, where we're dealing with scientific output that's not subject to copyright. This is just data coming out of experiments. What's important there is that we have common standards in terms of what kind of data we have, the metadata we have, how do we construct that? So copyright is less of an issue.
It's the balkanization of knowledge that is hindering our use, let alone all the issues around the regulation of AI. Open science is not one thing. It lies along a spectrum, from smaller groups that share knowledge internally and then license externally, all the way to very open uses where the data is made available to everyone at the same time.
And so there's no such thing as one open science. The Linux example is a particular one that comes out of software, which is not really open science, it's open source, it's a related concept but not the same thing. And open science is not aimed and will never be the cure to everything. It is one tool among other tools. Intellectual property is a tool.
If we're concerned, as you seem to be, about large companies taking control over certain knowledge, there's only so much we can do. We can do a lot through open science by making sure that SMEs, that researchers, that the Global South can participate. We need to worry about the rules of participation to make sure that we're not using open science as just extracting knowledge from the Global South and then selling them back products, but actually getting them involved. And we have a number of projects where, for example, we're supporting a low-energy MRI device that could be used in resource-deprived areas, that includes Canada's North but also many areas of Africa where we're working, where our researchers are working with researchers in the Global South. So there's no one thing, in two hours it seems like there's one approach to open science, but in fact there are multiple.
I know I only touched on answering a few of your points, but hopefully that contributes a little bit.
Thank you so much, Richard. I'm trying to get eye contact. Yeah. Well, there's another replier now from Professor Nilsonne.
Gustaf Nilssonne
Thank you, Katharina. Very briefly, I want to thank the questioner for bringing some more spirit into this debate. Sometimes it gets a bit abstract when we're talking about all these concepts, and I'm very curious, I will look up later what metacommunism means. But I do think that regardless of ideological standpoints, we can agree broadly speaking that knowledge is a basic human right, and personally as a scientist I am very embarrassed that we don't yet publish all our research outputs openly so that anyone, any citizen in Europe or in the world, can read and make use of this, of these research outputs, and even more so when it comes to research data, which is where the actual information is compared to research papers, which contain summaries, broadly speaking. So thank you for that.
Speaker 19
Thank you so much Gustav. I will I assume now this is closed thank you again and there's one and this will be our second and last question Stephen.
Stephen Wyber
Thank you very much so um should I stand up or okay I should stand up okay I've got my notes written down here. So um Stephen Wyber director for external affairs at the International Federation of Library Associations and Institutions long name so I say it quickly. Um thank you so much for all the interventions so far it's been a really good sort of mixed panel I think when we're looking at AI and and IP I think we're beginning to see some of the limits of trying to regulate purely around IP and especially around copyright as a tool there are there are gaps there are things it doesn't actually solve and so I think actually the shift towards the AI Act is a very positive thing we're actually trying to legislate for goals rather than specific tools and I think that that aligns also well with the simplification agenda it aligns well with what the IP4OS project has been doing by trying to turn a whole sort of thicket of different rules into a coordinated set of advice for researchers.
I suppose the question coming out of this is, in this world where we are starting to legislate much more clearly for the goals we want to see, what does this end up looking like for institutions themselves? How does this sort of simplification agenda translate into the rules, the procedures, the expectations that institutions themselves face?
Manuel Aleixo
Very good. I assume you meant research institutions rather than public policy institutions. So I'm not best placed to reply, but I will anyway, of course, because that's what the Commission does. The, I think, I think the objective here of having EU-wide rules on open science is precisely to have a common level playing field. Now, again, I emphasize the caveat, we are still asking ourselves, do we need the common level playing field?
Do we need EU-wide rules? Because we have to have considerations of proportionality here and of the impact of what we propose. But this is broadly the question.
Manuel Aleixo
And I think the generally the idea of having EU-wide rules on open science is that knowledge flows more freely throughout Europe. And this is exactly the purpose of the European Research Area, that somehow, with the mix of policies and rules that we put in place, we achieve a situation where it's easier and easier for researchers, for knowledge to circulate throughout Europe. And of course to ensure ultimately that Europe is a great place for research and for scientists to be and to develop their work. So this is ultimately the purpose. And this of course translates or has to translate in greater simplification and greater simplicity for all the actors in the research ecosystem, be they researchers themselves, funding organizations or research performing organizations.
So I think ultimately this is what we want to see and this is what we want to achieve, to move from an ecosystem that consists on a patchwork of different legislations and different practices, etc., to a more harmonious and EU-wide system that can allow a European Research Area and the fifth freedom to develop.
Lina Gálvez
Just very quickly, no precedent that I agree with the Commission.
But I do, I really do, so I will not add anything to what he has said. However, there is a risk. Because when we talk about simplification, I've been a researcher all my life until I was almost 50 years old that I came here to Parliament, so I've been, I spent more time in that side than in this side of the room. And when we think, or you think of simplification as researchers, mainly you think simplification for users, for you. But the agenda that it is developed is not necessarily this one.
Sometimes it's more deregulation than simplification. And deregulation does not bring us to this level playing field, this common rules for Europe. So this is the, yeah, the thing I would like to mention.
Speaker 19
Yeah, thank you so much to these replies. And I think this leads us now to the last point of the agenda, to the closing remarks.
Speaker 27
Dear Vice-Chair of the Panel for the Future of Science and Technology, Lina Gálvez, the floor is yours.
Lina Gálvez
So I've been taking a lot of notes. I hope I will be able to make a a conclusion that really brings to all of us what it has been said, the very, very, very rich discussion we we had and how timely it is. Obviously, many of the issues we have been dealing with here will be, I hope, developed in the ERA Act that we are really looking for here in Parliament, especially me, I have to say. And uh but um we are at that moment also discussing here at the European Parliament the new framework programme, the new the FP10, Horizon Europe. We will have our last shadow in two days, the 15th of October, and one of the the rapporteurs from my group, from the S&D.
So I have taken um many notes also for this process.
So coming back to STOA and to this workshop, this is not the first time we are discussing here in STOA open science. For instance, I decided to invite also Timo because he's more for the science part and he's more for the intellectual property part. So we thought it would be a good way. It is not the first time, as I said, we've been discussing here open science and even the relationship with artificial intelligence. But we wanted really to land it on also specifically with property rights in order to answer a very clear question.
So if our IP framework fit for open science within the artificial intelligence era? So if I have to answer, obviously, if I have to answer, probably the question is not. It is not. It is not.
Even if as we here, we have been taking the right decision, so doing the correct thing. I mean, it was Timo Wölken saying, we have been taking the right decision. It was Manuel just saying, we have been from the, from the intellectual property rights and from the science, open science, so far we have been taking the right decision. But as Professor Puigdemont was saying, science is going too fast. Everything is going too fast.
Artificial, well, too fast, fast, very fast. And artificial intelligence development too. So probably we really need this rebalance you were mentioning all the time in order we really face this knowledge valorization. It was, it is essential that we have the tools, the capacity building. I was really very interested about the awareness issue.
Because you were saying that it was a limited awareness, but it was also said that since artificial intelligence arrived, the awareness has been raising as well. And it was very interesting you were saying, well, that could lead to, if there is not enough awareness, that could lead to conservative answer. People will say, that's my data, my treasury, like in the Lord of the Rings, no, I will not share with anyone. Or to be risky.
I mean, just to share your data without knowing that you could lose intellectual property over it. But there is another possibility, another possible answer that in previous discussions in other workshops with and you know, being one of the rapporteurs of the Horizon Europe, you can imagine I receive a lot of researchers, a lot of associations and talking to a lot of people. And there are already scientists saying, I'm thinking to cheat in my data. I'm thinking to to to to put wrong data on my data. Like this, next time this mining engineers will use my data, I still remember a physics saying, they will use my data, they will get it wrong.
Because it's the data that took me a lot of time to build up in my lab and they are using without any context and and without any knowledge. And and this is something I would like to raise, probably for ERC and on that is that not all disciplines are in the same position. It's the same that not all disciplines work in English.
It's true that more and more is the case, but not all of them. Not all of them needs the same context or to be contextualized in each nation state or in each whatever. So this is something that could happen as well. So we have to to have that in mind as well as we know that there are artificial intelligence or people that are creating fake papers, fake research in order to feed artificial intelligence in the wrong way. So this kind of terrorist answer either from a researcher perspective, either from a really political fake news perspective, there is something we really need to have also in in in mind.
Anyway, um, it is as I say, it is definitely um a debate we can have when the AI Act will will come. In in any case, there is a I would say at the very beginning, this is a very timely discussion because there is certainly um a window of opportunity in the upcoming legislation, such as AI Act, but probably on the Innovation Act we can discuss um as well, or further initiatives addressing data. Regarding the the biotechnology, let's let's let's think about it that because it's not maybe it's not the best moment because it is true that you were also mentioning that we really need to align science policy, artificial intelligence governance, research integrity, copyright, research assessment, infrastructure policy, and I completely agree with that. But we really need to to align all that with the ongoing political moment. It has to be done within this the reason why probably this biotechnology act is not the best moment to to to to reopen uh because it's a moment where anti-science is reaching institutions.
A third of this Parliament is denying climate change or other issues. So this is the reality we have.
Lina Gálvez
We have that even in leader countries. So this is probably something we really need to rethink whatever we do, how it's going on in this current time and who's now owning this artificial intelligence tools. It's not publicly owned most of them. So all that we really need to rethink in that political moment. So I would like to finish having expressed a couple of concerns as you have seen.
I would like to express thanks to all speakers, especially to Katharina Miller for putting together this wonderful panel and acting as moderator, also to the audience also for your contributions.
It is also most appreciated and natural at the same time that we approach our topic from an international perspective. And I was really glad to have Professor Gold with us, not only because obviously science is open, is what we are trying to defend here, but also because Canada is a strategic partner, is associated to Horizon Europe and because of specific collaborative platforms as the digital partnership. So, this time of multiple tensions, geopolitical, environmental, socio-economical, there is no time to waste the knowledge and ideas generated by our scientists. But as we heard today, many solutions in the European Union exist within the existing rules, but yet we need to increase clarity and coordination. Sometimes with binding.
I was, I have, this is something the Commission and Parliament we are doing all the time, so don't get... I really hope that there is some kind of binding rules on the ERA Act regarding the research assessment, because that is at the end also a way to guarantee mobility and a way to guarantee many things.
I think this is too important for researchers and for good science that probably we really need to go beyond some guidelines. I mean, it's just a proposal. It's just that. So in any case, we hope that the ERA framework allows for creating and putting in place efficient and much-needed guidelines, and some of them binding, and toolboxes for researchers and knowledge management institutions. There is certainly, as we were saying, a window of opportunity, and I would like to thank you all of you again because this has been extremely useful, I'm sure for the Commission, but also for the Parliament in the coming regulation we are doing, that at the end is what STOA for.
I mean, to have this discussion knowledge-based with science knowledge and inputs in order we can take the best possible decisions also as policymakers for you, for the whole of the society. So thank you very much for coming. Thanks.
Ask anything about this video — I analyse its transcript to answer.