Skip to main content
Loading Explore guides…
Technology Ethics Principles · Acedra
Home
Explore
Computer Science
Technology Ethics Principles
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
155 segments
Copy
Hide video
0:05
Elizabeth Renieris: Hello everyone and welcome, thanks for joining. My name is Elizabeth Renieris, and I'm the founding director of the Notre Dame-IBM Technology Ethics Lab. Elizabeth Renieris: And we are so happy to host our inaugural spring event today on moving from principles to practice and technology ethics.
0:22
Elizabeth Renieris: The mission of the lab is to promote human values and technology, with the ultimate aim of helping steer technological design, development, deployment, and use in the direction of more equity and justice. Elizabeth Renieris: We do this by producing and supporting tangible applied and interdisciplinary research projects to address core ethical questions,
0:39
Elizabeth Renieris: convening facilitating dialogues and debates around technology ethics, and working to develop an informed permanent standards laws and policies related to the moral and ethical use of technology. Elizabeth Renieris: Turning to today's event, the rapid spread of artificial intelligence and related technologies in recent years has been accompanied by a proliferation of principles-based ethical frameworks intended to guide their design, development, and deployment and practice.
1:03
Elizabeth Renieris: According to the UN Secretary General's Roadmap on Digital Cooperation, there are more than 160 distinct organizational national and international sets of AI ethics and governance principles worldwide, Elizabeth Renieris: and countless others related to technology ethics, more generally, but despite the existence of hundreds of sets of principles we have limited evidence of their impact and practice.
1:27
Elizabeth Renieris: In today's panel, we will explore efforts to translate principled frameworks into principled action through a cross-sectoral interdisciplinary dialogue in line with the Lab's mission. Elizabeth Renieris: I'll begin by briefly introducing each of our distinguished guests today. Their full bios can be found on the event website, and going through them would keep us here all day.
1:46
Elizabeth Renieris: After that, I'll lead us in a discussion for about 30 to 40 minutes, leaving enough time for pre-submitted questions from the audience, and for those who registered and submitted questions we do thank you. Elizabeth Renieris: So first up, we have Jessica Fjeld. Jessica is a lecturer on law and the assistant director of the Cyberlaw Clinic at the Verkman Klein Center for Internet and Society at Harvard.
2:06
Elizabeth Renieris: Her work focuses on legal issues impacting digital media and art, including intellectual property, freedom of expression, privacy and related human rights issues, as well as contract and corporate law. Elizabeth Renieris: Recently, she has focused on AI-generated art, the overlap of existing rights and ethics frameworks, and emerging technologies and legal issues confronted by digital archives.
2:27
Elizabeth Renieris: She's a board member of the Global Network Initiative, GNI, a multi-stakeholder organization that protects and advances user freedom of expression and privacy around the world. Jess, thank you for being here. Elizabeth Renieris: Next up, we have John. Havens. John is the Executive Director of the Council and extended intelligence and IEEE Global Initiative on the Ethics of Autonomous Intelligent Systems.
2:49
Elizabeth Renieris: He's also the founding and current vice chair of the IEEE 7010 Well-Being Metrics for Autonomous and Intelligent Systems working group, Elizabeth Renieris: the founding chair of the IEEE P7000 Draft Model Process for Addressing Ethical Concerns During System Design working group, Elizabeth Renieris: and founding co chair of both the Personal Data and Individual Access Control Committee as well as the Well-Being Committee of the IEEE's Global Initiative.
3:10
Elizabeth Renieris: John is also the author of Heartificial Intelligence: Embracing Our Humanity to Maximize Machines and Hacking Happiness: Why Your Personal Data Counts and HowTracking It Can Change the World. Elizabeth Renieris: And I believe he's also the only professional actor joining us today. John, thank you for being here. John Havens: My pleasure, thanks for having me.
3:29
Elizabeth Renieris: We are also joined by Philippe-Andre Rodriguez. Philippe is the Deputy Director of the Center for International Digital Policy at Global Affairs Canada, Elizabeth Renieris: the Canadian government's department in charge of diplomatic and consular relations, as well as a professor of the practice at McMaster University. Elizabeth Renieris: He previously served as senior advisor at Canada's Privy Council Office on issues related to digital platforms and AI governance.
3:52
Elizabeth Renieris: Prior to joining the Canadian government, he was a fellow at Yale Law School and completed a PhD at the University of Oxford as a Rhodes Scholar. Elizabeth Renieris: His research has been published in the Cambridge Review of International Affairs, European Review of History, and Human Rights Review, among many others. Philippe, thank you for joining us.
4:09
Elizabeth Renieris: And next up, we have Francesca Rossi. Francesca is an IBM fellow and the IBM AI Ethics Global Leader, and she leads AI research at IBM T.J. Watson research lab in New York. Elizabeth Renieris: Francesca obtained her PhD in computer science from the University of Pisa and taught computer science both there as well as at the University of Padova for nearly 20 years before joining IBM. Elizabeth Renieris: She has published over 200 scientific articles, as well as a book in 2019.
4:37
Elizabeth Renieris: Francesca is a global leader in AI ethics, the founder of the IBM internal ethics board and a leading contributor to many global initiatives around this topic. Elizabeth Renieris: She has been a member of the European Commission High-Level Expert Group on AI, part of the Global Partnership on AI Coalition, and she will also be the next President of the Worldwide Association of AI. Francesca we're so glad to have you today.
5:00
Elizabeth Renieris: And last but certainly not least, we are joined by Fabro Steibel. Elizabeth Renieris: Fabro is a postdoc affiliated with the Berkman Klein Center for Internet and Society at Harvard University and a member of the Global Council of the World Economic Forum. Elizabeth Renieris: For six years he served as an independent researcher at the Open Government Partnership in Brazil.
5:18
Elizabeth Renieris: And he's also a fellow in open government with Organization of American States, a postdoctoral fellow at the United Nations University, and visiting researcher at the University of California San Diego. Elizabeth Renieris: Fabro is the executive director for the Institute for Technology and Society in Rio, a leading civil rights, excuse me, leading civil society organization in Latin America.
5:38
Elizabeth Renieris: So Jessica, John, Philippe, Francesca, and Fabro, thank you all so much for being with us today. Without further ado, let's dive right in. Elizabeth Renieris: Jess, I'd like to start with you. You were a lead researcher and author on Berkman Klein's Principled AI project and report. Elizabeth Renieris: which sought to map emerging areas of consensus across the many ethical and rights based approaches to Ai principles.
6:02
Elizabeth Renieris: Based on your research, could you please tell us more about the project and its findings share your view on the state of Ai ethics and it's normative foundations and perhaps also highlight any shortcomings of existing approaches or processes. Jessica Fjeld: Absolutely, thank you Elizabeth and hello to my fellow panelists i'm delighted to be here with you today i'm just going to share my screen. Jessica Fjeld: Can you all see that great i'm so.
6:30
Jessica Fjeld: happy to talk happy to be. Jessica Fjeld: here to talk about the principal day I project in some ways it's a little bit of a snapshot in time, but it was certainly. Jessica Fjeld: A response to the kind of feeling of overwhelm that came on us as we were watching quite a number of those 168 principles documents that Elizabeth referenced at the top of the program trickle in.
6:52
Jessica Fjeld: As we were seeing that happen and we are running the Ai ethics and governance initiative at berkman Klein, we were really starting to think you know. Jessica Fjeld: Apart from reading these documents sort of one by one, as they come in, how could we arrive at some kind of yardstick to understand how these documents relate to each other.
7:12
Jessica Fjeld: We can tell that they have some things in common What are those things where they diverging and where we landed was was the principal the I project, which was published in January 2020 so it's a little more than a year old now. Jessica Fjeld: But I will go, we go. Jessica Fjeld: So it is a White Paper, and it is also accompanied by this data visualization which I think is to me really helpful and actually was a core goal of the project, even as we were compiling the research.
7:44
Jessica Fjeld: What we did was take a look at the breadth of principles documents principle, you know, Ai principles documents that were available at the time. Jessica Fjeld: That we were doing the research and 2019 and try to curate a data set that was small enough to be manageable but broad enough to represent kind of the conversation at the time, so. Jessica Fjeld: This in this visualization the every spoke of the wheel is one set of Ai principles, with the exception of the OECD and.
8:18
Jessica Fjeld: principles which, apart from some preparatory material are functionally identical and their wording so they actually appear on one spoke. Jessica Fjeld: Each color corresponds to a stakeholder groups, so you have government in green intergovernmental organizations in orange multi stakeholder groups in blue private sector and pink and civil society and yellow.
8:40
Jessica Fjeld: And then each ring of the wheel corresponds to one of the themes that we arrived at so just to get a sense of of the body of data, we were looking at at the time we published this, I think there were more like 100 principle document principles documents out there. Jessica Fjeld: This shows the documents that we chose to include on a timeline those colors are the same, so you can see that the the earliest documents.
9:08
Jessica Fjeld: There are a couple of multi stakeholder documents for the partnership, and I am Ai and sylmar as well as some government documents early on, but the bulk of the documents we were looking at were coming in over 2018 and 2019. Jessica Fjeld: We have Filipe Andre on this call, so I will note that in scoping this we were looking directly at.
9:32
Jessica Fjeld: principal statements so, for example, at this time, Canada had adopted its world leading regulation on the government acquisition of Ai technology, we did not include that document, although we were aware of it and meyers of it. Jessica Fjeld: Because it was adopted as a regulation and we wanted to compare head to head really principles documents. Jessica Fjeld: So what are some of the themes that we arrived at. Jessica Fjeld: A we have.
10:05
Jessica Fjeld: From the Center out and they're in no particular order here actually they're arranged in the order that they are mostly because of how much space, the words took up on in the visualization itself so privacy is the tiniest one and fit in that middle ring. Jessica Fjeld: So we have privacy accountability, safety and security, transparency and explain ability, fairness and non discrimination human control of technology.
10:35
Jessica Fjeld: Professional responsibility promotion of human values, and then the outermost ring collects information on whether the document in question. Jessica Fjeld: mentioned international human rights either just in terms like effectively in those in so many words if it's just a diamond if it's a star it referenced actual instruments of. Jessica Fjeld: international human rights law and referred to itself as sort of proceeding from an international human rights framework.
11:04
Jessica Fjeld: The size of the.at the intersection of each spoke and ring indicates how many principles that fit into that category, the document covered so there's a huge range in terms of the principles. Jessica Fjeld: Of hang on, I have a slide here, somewhere, here we go. Jessica Fjeld: In terms of how many principles fall under each theme, so you can see, for example, there are quite a number of principles under accountability.
11:33
Jessica Fjeld: ranging from accountability writ large, to recommendations for new regulations impact assessments auditing requirements verify ability and replicability liability. Jessica Fjeld: Concerns appeals processes environmental responsibility, creation of a monitoring body and remedies for automated decision making. Jessica Fjeld: And you can Contrast that with you in control of technology or promotion of human values, the ladder has just three principles within it.
12:01
Jessica Fjeld: Whether the Ai and question is leverage to benefit society, whether it is used to promote human values in her miss human flourishing and whether there's a question of kind of access to technology in the Ai principle statement. Jessica Fjeld: So I mentioned that, because when we look at these rings it's important to remember that you can compare within the ring So you can see, for example. Jessica Fjeld: that here we go this is.
12:30
Jessica Fjeld: The Dubai Ai principles and ethics document, and this is the German Ai strategy that, in fairness and non discrimination, the Ai sorry fairness and automation is here the Dubai document covers more of those principles than the German document does that mean it is better now. Jessica Fjeld: It just has a broader breadth of coverage, and so I think the thing that we we took away a number of things. Jessica Fjeld: From this research.
13:03
Jessica Fjeld: At the end of 2019 and into the beginning of 2021 was that there really was this sort of convergence of themes and I was reassuring when we publish this research, there are a number of other. Jessica Fjeld: sort of similar papers that came out in the month immediately before in the month following and while they group we all grouped themes very slightly differently, I think that these kind of core themes.
13:24
Jessica Fjeld: Which hadn't been identified before were largely apparent across those papers. Jessica Fjeld: And so I think we did start to see that there was a kind of conversation emerging in the governance field that may feel in 2021 obvious. Jessica Fjeld: It certainly wasn't at the time we were doing the research and the thing that I think is the bigger challenge, and the reason that I want us to put a little bit of an asterisk around that.
13:55
Jessica Fjeld: Consensus is that I mentioned at the top, that we did purpose of sampling here right, so we didn't this was not an effort to include every principles document that we could find, but rather to sort of come up with a strong sample. Jessica Fjeld: That would represent the breadth of the conversation, and you can see in this chart that breaks down by stakeholders that we were able to get a pretty good breadth of representation.
14:18
Jessica Fjeld: You know, from a lot of government principles, yes, but like decent number of civil society, private sector multi stakeholder principles as well i'm going to skip sorry these slides are there, we go, this is what I want. Jessica Fjeld: When you we also strove really hard to find geographic diversity within the principles documents that we included, even though we had a vastly multilingual research team.
14:44
Jessica Fjeld: With speakers of something like six or eight languages and we're researching in those languages in addition to you know just an English, which I speak leading the team. Jessica Fjeld: You see that this is dominated by North America and Europe So yes, we can say that this research revealed a kind of a core. Jessica Fjeld: set of themes, but also, it is a theme is a set of themes that are highly dictated by the.
15:15
Jessica Fjeld: governments and private companies and civil society organizations that tend to be best resource and have the loudest voices in North America and Europe and it's done and to some extent East Asia other especially China. Jessica Fjeld: Their vast regions of the world and regions of the world that are likely to be highly impacted by Ai that are just really not represented here.
15:39
Jessica Fjeld: So I think that that's an important caveat to kind of take with us as we start this discussion, which is not only. Jessica Fjeld: How are we doing at implementing the principles that we have, but also what is the foundation of those principles and is that the appropriate foundation to govern Ai and i'm going to wrap up there. Elizabeth Renieris: Thank you so much Jeff that was.
16:00
Elizabeth Renieris: A really wonderful overview, I think a really great place to start this conversation and take stock of where we are. Elizabeth Renieris: I found the pie charts particularly helpful and illustrating the participation of different sectors, as well as different geographies.
16:15
Elizabeth Renieris: and trying to bring in different sectors and geographies was really important to this conversation I do want to appreciate the difficulty with the pandemic and that different parts of the world are really impacted in very profoundly on even ways. Elizabeth Renieris: So I do appreciate everyone being here that said it's also interesting to notice how the consensus, you talked about has some of the shortcomings we've seen.
16:36
Elizabeth Renieris: For example, and human rights frameworks which also have been criticized for not necessarily representing a global consensus, but over representing certain parts of the world, and under representing others so i'm sure we'll turn back to some of these themes. Elizabeth Renieris: Right now i'd actually love to pivot and turn to Francesca so.
16:54
Elizabeth Renieris: Francesca you know, given your own involvement in many of the organizations and multi stakeholder groups that were actually mapped in that principled Ai project. Elizabeth Renieris: As well as your many roles that IBM was hoping, you could tell us about how companies like IBM or translating some of these principles into practice.
17:10
Elizabeth Renieris: Including how you leverage some of the foundations that just outlined, for us, and any challenges that you faced in turning these principles into principle to action, particularly within the structure and hierarchy of a large global Corporation. Francesca Rossi -- IBM: Yes, thank you Elizabeth. Francesca Rossi -- IBM: Let me share my. screen. Francesca Rossi -- IBM: Okay So yes, so um, so I think that.
17:37
Francesca Rossi -- IBM: The whole discussion around Ai ethics and how to operationalize the principles, whether it's in a country or in a company has to start with the vision of what one wants to do with this technology. Francesca Rossi -- IBM: And and usually I call it the vision of but actually more with Ai, meaning that I mean the goal here is not really to advance the ui per se, but to.
18:02
Francesca Rossi -- IBM: advance people societies, the way we live our values with the technology and so how, how can what would be the properties of these technology that can help us reach that vision, and I would say that the you know, Ai is still not. Francesca Rossi -- IBM: General and adaptable enough for so Ai as a lot of success in.
18:26
Francesca Rossi -- IBM: Solving many different problems, but he still has limitations that that's why there is a larger research community that tries exactly to address those limitations. Francesca Rossi -- IBM: To make it more trustworthy you know reliable robust explainable aware of its bias and so on respecting privacy, human rights and many of those. Francesca Rossi -- IBM: sites that already earlier on, to make it more able to work with us so supportive and collaborative.
18:53
Francesca Rossi -- IBM: We do, will be able to work with human beings and to support our decision making capabilities, which are also limited in some sense and other other says, maybe we are even better the machines, but we are complimentary. Francesca Rossi -- IBM: And then also to make it sustainable a competition and his character, so I think that maybe.
19:11
Francesca Rossi -- IBM: I mean, we need to have a vision of what will want to do with Ai to be able to understand what are the principles and then how we operationalize these principles so when. Francesca Rossi -- IBM: I present a is Maggie presentation around the it, so I always have a slide with the principles of a project and they around visual that.
19:33
Francesca Rossi -- IBM: Just presented earlier, so I didn't put it here this slide because I know I knew that jess was going to go into details, but I really. Francesca Rossi -- IBM: Like the the way it was conveyed in ways was put together but also put it there, to say that that yes, we have enough principles we are done with the principles, now we have to move forward from the principles, but when I think about the principles.
19:59
Francesca Rossi -- IBM: I also like to think about because I received this question and people say why Ai ethics, so why not technology ethics in general, why so What is so special about Ai so. Francesca Rossi -- IBM: Here I just played down some of the issues that are also in in what just presented, but I try to relate it to some features of a yeah I like the needs of a large amounts of data.
20:25
Francesca Rossi -- IBM: The fact that is often these techniques currently are kind of a black box so that's why explain abilities initial responses. Francesca Rossi -- IBM: The fact that they I can make all recommend decisions, so we have to make sure that the values in making the decision of the same like feminists as we like. Francesca Rossi -- IBM: The fact that is based on statistics isn't always a small percentage of error if you use machine learning so that's why.
20:48
Francesca Rossi -- IBM: people think Oh well, then what he was accountable if there is that error, because after all that ever doesn't come from an arrow from some developer right. Francesca Rossi -- IBM: And then other ones, so I mean, I think that there is still some discussion to do about principles in how to relate them to the characteristics of these technology, particularly.
21:07
Francesca Rossi -- IBM: But I think, really, we need to move on, so the principles for me are just the beginning of a long journey. Francesca Rossi -- IBM: that a country or a company has to do in order to understand really how to change things, how things are done how Ai is the design develop deliberate used, and so on. Francesca Rossi -- IBM: So at IBM we started with our principles very simple very high level.
21:31
Francesca Rossi -- IBM: Their core principles of trust and transparency 2013 and we said well first, we think that Ai should among human intelligence support the human. Francesca Rossi -- IBM: Decision making second one is about principle is about the data that belonged to whatever provides the data and the third one, is a focus on transparency and explain ability.
21:50
Francesca Rossi -- IBM: And you see that here we talked about principle of trust and transparency, but we don't really say what trust means. Francesca Rossi -- IBM: So the next thing that we did the after publishing this principle was to say what we mean by trust so trust can be many different things at IBM we said let's focus on for things.
22:08
Francesca Rossi -- IBM: Besides accuracy, of course, and besides, privacy, the four things that we want to focus on is fairness is for inability robustness and transparency, so we gave a structure to this notion of trust. Francesca Rossi -- IBM: Which is not just trusting the technology but trusting the whole ecosystem that produces and uses this technology.
22:29
Francesca Rossi -- IBM: So then, once we had these two things it was still not enough, because of course these tools lights are not enough to give the developers and say okay now change the way you produce a ui that's. Francesca Rossi -- IBM: Not what that what is useful to them so that's why we put together a lot of other activities guidelines toolkits a lot of educational material culture change design thinking sessions. Francesca Rossi -- IBM: A lot about the governance of being able to.
22:59
Francesca Rossi -- IBM: supervise and coordinate of these activities, and then a lot of work with our clients or partners or entire ecosystem and also all the partnership that we have with the academia, like this one, with Notre Dame but also with other companies other. Francesca Rossi -- IBM: Institutions organizations and salt and you see that here, I put this pinker I was saying that this is not a static thing this is evolving, this is a journey that the company has to go through.
23:27
Francesca Rossi -- IBM: In putting together all these pieces so for example at IBM as a governance centralized governance, we have put together this Ai ethics board that coordinates all the activities around the. Francesca Rossi -- IBM: Ethics, and this is especially important, the coordination in a in a large company like IBM, a lot of. Francesca Rossi -- IBM: activities to change the way people think about these issues, no education awareness, for developers for managers for the executives.
23:55
Francesca Rossi -- IBM: As well as technical solutions like open source or proprietary tool kits to the tech bias and mitigate it to provide a sprint ability, transparency and so on. Francesca Rossi -- IBM: As well as a lot of help for our developers to change the way they produce Ai and make sure that they include these Ai ethics, the handling the tools and methodologies integrated into their development pipeline.
24:25
Francesca Rossi -- IBM: On the left you'll also see these four areas which is, which are the four areas in which at a ui researcher at IBM is structure and and you see that one of the four areas is about yesterday yeah so we so a lot of. Francesca Rossi -- IBM: Initiative come from pushed from research, because there are maybe innovative results in our team algorithm that then are pushed into our open source of propriety toolkits to be used.
24:53
Francesca Rossi -- IBM: Or you also see this Ai fairness 360 is credibility 360 which has to open source tool kits that we put together for these two issues, as well as these. Francesca Rossi -- IBM: booklets that we put together for our designers and developers is called every day is for a yeah that tells them what we're running examples, how to check how to be more aware of their own biases and what it means so for a system to be explainable.
25:19
Francesca Rossi -- IBM: A particular emphasis, we put on the issue of transparency through this mechanism that is we call it the fact sheet. Francesca Rossi -- IBM: Which is basically one way of implementing this idea of transparency by documentation restricts developers and designers need to respond to these long list of questions that allow them to. Francesca Rossi -- IBM: document all the design choices that they make during the development and design the pipeline.
25:47
Francesca Rossi -- IBM: But it's important that is not just a checklist to be filled, at the end of the development process but it's something that should be. Francesca Rossi -- IBM: considered throughout the whole pipeline and is useful for themselves as well to change.
26:01
Francesca Rossi -- IBM: Our new processes, if needed, but also for clients, especially those that work in a very regulated the domain, but also for users for regulator for all the tours so that possibly want to know what happens when developing that the Ai system.
26:17
Francesca Rossi -- IBM: Because i'd like to hear so to the say a little bit more about the ethics board, as I said that a big role is awareness and coordination, but also supervised internal education and retraining of various roles within the company around the it. Francesca Rossi -- IBM: links the research division for the services into the platform and product division so which otherwise it would be.
26:40
Francesca Rossi -- IBM: Less seamlessly linked to each other it's also gives advice to the offering teams to make sure that the what they offer what they deliver to our clients. Francesca Rossi -- IBM: is aligned to our principles and we use a risk based approach, where we vet the technology itself the uses of the technology ends well as the client and the context in which the technology.
27:02
Francesca Rossi -- IBM: will be used, we developed new methodologies for, as I said, for developers designers. Francesca Rossi -- IBM: We have internal governance framework for each one of these activities we work with our clients to define a sort of a is interoperability, to make sure that our clients understand what we accept, and they work with the. Francesca Rossi -- IBM: Principles and with open nationalization of principle that are compatible with our words.
27:26
Francesca Rossi -- IBM: And then we also define policies and inform or advisor regulators here is a picture, with all the partnership model, many of the partnership that we are but. Francesca Rossi -- IBM: The main point here is that is very multidisciplinary very multi stakeholder with academia companies government civil society organization.
27:44
Francesca Rossi -- IBM: And the idea here is that we bring our expertise into this partnership, or we also learn a lot from all the stakeholders that we we work with inside these partnerships. Francesca Rossi -- IBM: And so we think that no single company, no matter how big it is or how all the things like IBM can do it by itself, you know, Ai is around now by now definitely a multi disciplinary a multi stakeholder for.
28:12
Francesca Rossi -- IBM: One thing that is very recent that I wanted to add here is the new proposal for the Ai regulation that was published by the European Commission, just a few days ago. Francesca Rossi -- IBM: Last week, and and here, this is very aligned that we want to we have in mind about what I go letting go yeah and, of course, the regulation around the Ai isn't one more way of.
28:34
Francesca Rossi -- IBM: going from principles to practice, you know, the European Commission had to list, for example, all the. Francesca Rossi -- IBM: Ai applications that are acceptable or have a high risk, low risk of minimal risk and then give obligations to many stakeholders, for each one of these levers and. Francesca Rossi -- IBM: The point that is nice here that we really welcome and that we have also been advocating, in the past.
28:58
Francesca Rossi -- IBM: Is this idea of procedure regulation which is basically the fact that first you regulate and Ai applications and not the technology per se. Francesca Rossi -- IBM: And SEC on that you have a risk based approach in this particular case, there will be a Commission chose for lenders of risks. Francesca Rossi -- IBM: And and and then the emphasis on building trust you know that is also so central for us.
29:21
Francesca Rossi -- IBM: But also we see it also in this proposal, this proposal has been based on many consultations, of course, but i'm very happy to see that the work that we did. Francesca Rossi -- IBM: With the high level expert group is really present in this proposal for regulation in many different ways, the issues, the risk based approach.
29:42
Francesca Rossi -- IBM: and many others, so the transparency emphasis the human oversight and so on, and the last bullet point here is that also this revelation proposal. Francesca Rossi -- IBM: acknowledges the fact that even regulating the ui is a place where many stakeholders have to play their path providers users and many other stakeholders so in the proposal that our obligations, not just on those will provide a yeah.
30:09
Francesca Rossi -- IBM: But also the users and also others and that's something a very important point to be understood so here I just put some useful links for people and i'm happy to share this life, thank you. Elizabeth Renieris: Thank you so much Francesca a very, very clear articulation of what it means to turn these principles into practice. Elizabeth Renieris: And it really excellent point as well on regulation as a form of opera operationalize the principles.
30:37
Elizabeth Renieris: Another really tangible operational ization of these principles, has come through standards and Standards Organization so with that I would love to turn to john. Elizabeth Renieris: john you wear so many hats I Tripoli, it can be hard to keep track sometimes. Elizabeth Renieris: But fundamentally I like to look at how standards bodies like yours are leveraging some of these principled foundations that we've mapped and outlined.
31:02
Elizabeth Renieris: And particular challenges and turning them into turning these normative frameworks into things like technical standards but also non technical standards which may be harder to measure or may pose more qualitative. Elizabeth Renieris: Challenges and standards so things like your work around the well being standards come to mind, but if you could just situate the standards community in this.
31:24
Elizabeth Renieris: Multi sectoral interdisciplinary a geographically diverse conversation around technology ethics and that's probably the best starting point, their. John Havens: journey no I have seven minutes to speak, but you really my title took four of them so sorry about that i'm actually i'm going to share my screen and I love I love the title that you chose for today let's see if I can get to it easily, how do we get to.
31:49
Elizabeth Renieris: Yes, it was in part, inspired by the ieee is ethics and action. John Havens: yeah well I just if I can get to buys it now let me share my screen. John Havens: Okay, maybe, can you go to that link easily Elizabeth so people don't just watch me search here. John Havens: i'll give it a shot here we go, this is my entire time is me looking to get to the right link anyway i'll started talking um so first of all it's an honor to be here and congratulations on the launch and also.
32:19
John Havens: want to point out that i'm honored to be here with the different guests, that you have on, and here we go thanks for making this. John Havens: There we go okay so ethics ethics and action is a website that we created about three years ago. John Havens: Where i'm going to start with the foundational work that we did called ethically line design now here's the thing about jessica's work that she knows we're friends and we talked about this.
32:42
John Havens: The work that she did is foundational and so helpful because comparing so many different documents God bless you I couldn't have done that JESSICA that so our. John Havens: What we did with this project is back in 2016 we did the first version of this paper, and we have people like Francesco see who is currently still on our executive committee who helped drive the work we had this amazing.
33:03
John Havens: event, and I remember really getting to know Francesca at that event in The Hague back in 2016. John Havens: Francesca was actually one of the first people we we called I think we said, like issues or problems or something we were in Francesca said wait. John Havens: Why are these these are issues to discuss, so this document is is was kind of designed to be a little bit different than other principles documents.
33:25
John Havens: Anyway, and also, I was new to the space when we started, but I bring this up to say we did the first version that came out in 2016. John Havens: We were honored to hear that that first version and second version. John Havens: we're hopefully helpful to people, but just like JESSICA pointed out our first version, it was extremely if not exclusively, North American and European.
33:46
John Havens: So one thing that we just got lucky and I Tripoli is focused on this in terms of how we do a lot of our processes, in fact, all of them. John Havens: is designed to be consensus driven and to your point Elizabeth about standards where where you create this is what I Tripoli does is sort of.
34:02
John Havens: As tools resources staff, like me, is to bring people together and, like all of us on the call share this vision is to cross pollinate across different audiences policymakers, academics, the business community. John Havens: To say how do we put these principles into practice.
34:17
John Havens: So that said after we get the first draft of ethically line design a lot of the feedback we got and I remember, I read through it 500 pages was people saying this is great stuff but it's really Western. John Havens: Now I bring that up because a lot of what we're talking about today and in terms of standards and you mentioned technical.
34:35
John Havens: With the idea of either socio technical or normative and Francesca just mentioned she's foundational and creating the EU Ai guidelines as part of a high level experts group. John Havens: it's fascinating to think about language, by the way right in terms of what is technical mean right or what is, and then, what does.
34:55
John Havens: You know how, how are we defining things that may not be technical anyway, I bring this up, because if you look at the chapters that are in ethically line design. John Havens: I was a part of writing some of these but there's about 700 people that wrote this document over the course of the three years in the three versions that came out.
35:12
John Havens: And I also really excited, for instance, the lead, one of the committee's etc, but there's a couple of chapters, I want to point out the relevant Elizabeth to your question. John Havens: One is the classical ethics and Ai chapter and Jessica knows about this work. John Havens: Because of other stuff we're working on together, this was a chapter that came directly from feedback.
35:29
John Havens: From people saying this feels where the Western I was like I don't know what that means, so we invited people to be a part of the committee in that chapter is written almost exclusively by non Western. John Havens: Experts saying what are things like the shinto tradition tradition in Japan, a boon to ethics from South Africa what our confusion oriented traditions.
35:49
John Havens: Why this is critical, is that, like I Tripoli, which is a global what's called an seo standards Development Organization. John Havens: A big part of understanding any aspect of this work is to ask the question, who isn't at the table, yet indigenous population certainly diversity in terms of.
36:07
John Havens: Women race, etc, so another thing I wanted to point out, and then i'll get more of the standards is well being as you can mention to you that word here and then also a personal data and individual agency. John Havens: In terms of the principles that are the first chapter of ethical line design consensus driven wise our teams voted on what those principles should be in order of importance which is really interesting.
36:35
John Havens: Meaning that's what the decision ones, human rights, like JESSICA was talking about that common theme. John Havens: One thing also we had our second big meeting in Austin in 2017 and a lot of the lawyers were like Hello ethics is great, what about international human rights law, and I was like cool. John Havens: So I bring that up to say when you design is amazing technologies, the order of how you think about what you think about is beyond critical.
37:01
John Havens: Right, if you are, at the outset of design saying this is our primary recommendation, but definitely line design and with responsible Ai or technological innovation. John Havens: is to say if I want to build this phone and i've already built say a $3 million prototype and then I start asking things about like transparency and whatever that's a very different set of constraints.
37:21
John Havens: and saying at the outset of the very beginning of design two things one is how do we make sure we're thinking about aspects of all these areas of risk and the end users. John Havens: But then the well being I just want to put out because it's critical well being is not about in our. John Havens: ethically wind design about mental health and physiological health alone it's actually an economic term.
37:42
John Havens: Because, when we talk about what we're talking about today a critical question to your point Elizabeth about well being. John Havens: Is what are the metrics of success or in the business parlance key performance indicators of what we're building. John Havens: and Francesca mentioned risk it's fascinating to think about risk and the EU has done some.
38:02
John Havens: groundbreaking work in their their report that just came out last week Similarly, the FTC in the United States and what suits them the same. John Havens: Which is risk being defined as physical risk well this autonomous vehicle crash, based on the lidar sensors incredibly important to ask. John Havens: But also asking us to you is done recently, what about how algorithms might score someone differently if they're from a marginalized population.
38:26
John Havens: in such a way that, then they they're just your sentences would be different than someone else right all i'm saying here is how we define risk. John Havens: In this present state regulation wise and with technology, etc, and how these are put into standards and law. John Havens: is evolving in ways, where we can't necessarily say this is technical, this is socio technical.
38:49
John Havens: Which is why this work led to a creation of standards, the standards series called the P, as in Peter 7000 series and that just stands for project. John Havens: which was really about saying how can we do mirror of the work that was done with about 3000 people who up i'm sorry 700 people they'll write this paper and about 3000 people in the larger community. John Havens: And like we're all doing here because we're all part of the same community.
39:12
John Havens: is to say, how do we take these amazing technologies and recognize, first of all the data agency is critical right we talked about Ai. John Havens: it's a big space it's machine learning it's all these different areas of study. John Havens: So much of it, if not all of it comes back to how is data access and in terms of our human data if we don't have access to the data. John Havens: and portability of that data things like being able to use blockchain or going to countries like.
39:38
John Havens: Finland or Estonia, who are really giving their citizens the opportunity to access their data and really have parity, but how that data is shared. John Havens: that's what we talked about an ethical line design built on top of gdpr things like governments and businesses. John Havens: being required to follow certain guidelines about privacy great but that's kind of like the outside in inside out is also data agency for human rights and integrity and agency.
40:05
John Havens: And the wellbeing already mentioned but they're the basic logic is sort of a triple bottom line people planet and prosperity. John Havens: We must be asking ourselves on this community is what we're building these amazing technologies, the primary constraint are measured success largely around one set of existing.
40:24
John Havens: Coming from GDP or shareholders are sort of exponential growth values even saying things like Ai for good building a technology that might eradicate cancer fantastic but is that all happening. John Havens: On the structure of the economics that exists now, the status quo, that means it's not that it's evil or unethical it's critical to come from a more empirical lens and say this economic.
40:48
John Havens: And if the constraints that businesses are put under our exponential growth, then, even when we are trying to protect the environment for human well being that comes after. John Havens: That first set of constraints are satisfied, and that means that, by definition, we are D prioritizing planet and people. John Havens: So, countries like New Zealand Scotland Iceland are prioritizing this triple bottom line and, by the way, it's also interesting they'll wrap up my seven minutes.
41:16
John Havens: that the EU is so progressive in terms of things like having health insurance for their citizens, comparing the EU and the states is often quite interesting and then comparing all of global North global South as interesting as well.
41:28
John Havens: Here i'll just say i'll put my information in the chat all of our standard groups are open, free to join would love to have you join and the standards are really the instance negation of this word from ethically line design, which is really putting principles into practice thanks. Elizabeth Renieris: Thank you so much john and we will be sure to also share those resources with folks and either tuned in are registered and.
41:49
Elizabeth Renieris: So, having looked at the perspective of the private sector and technical standards bodies, I do want to turn to the government and civil society. Elizabeth Renieris: And let's turn to Felipe next so with your background, both in academia and now government. Elizabeth Renieris: How have you seen governmental and public sector stakeholders like global affairs Canada.
42:08
Elizabeth Renieris: translate some of these principles into practice, we heard just mentioned how Canada was in some ways ahead of the curve going straight to Regulation before others were even working on principles. Elizabeth Renieris: And so, in that sequence How did the principles in turn inform some of the practice.
42:23
Elizabeth Renieris: And also feel, please feel free to highlight any specific challenges you faced and operationalize these frameworks, specifically and turning them into actionable diamonds for public sector interested stakeholders. MS Canada (Philippe-Andr? Rodriguez): Thank you, thank you so much, I was a bit, and you know this is a great panel and very, very timely actually this week I am.
42:45
MS Canada (Philippe-Andr? Rodriguez): Leading the Kenya delegation to the UNESCO recommendation on the ethics of artificial intelligence which is really the first entirely global process on the ethics of vr from a multilateral perspective so really timely. MS Canada (Philippe-Andr? Rodriguez): panel indeed so.
43:03
MS Canada (Philippe-Andr? Rodriguez): I don't want to take too much time, I see I see we are relatively short but I want to make a few points, first I want to go back to 2017 so I know we heard from john in terms of you know a lot of work was actually done. MS Canada (Philippe-Andr? Rodriguez): In the private sector and the standards world prior to, but for us the the story really starts in 2017 in Canada when.
43:25
MS Canada (Philippe-Andr? Rodriguez): at the global level, and I really want first to focus on on some of these global conversations and then I can I can speak to some of what we're doing in Canada, but we introduced the idea of. MS Canada (Philippe-Andr? Rodriguez): We need some kind of conversation on Ai between countries in a more sustained manner, given that we're starting to see some of the harms that are associated with.
43:47
MS Canada (Philippe-Andr? Rodriguez): The use of artificial intelligence and that's been really one of the drivers really to bring together different countries around around this this question and, at the time actually we brought this in the context of the G seven presidency.
44:03
MS Canada (Philippe-Andr? Rodriguez): And really we were quite surprised, but actually most countries we're not really thinking about artificial intelligence as a policy field outside of really something that is essentially an industrial policy, so they were not thinking necessarily about the harms and the ethics of.
44:20
MS Canada (Philippe-Andr? Rodriguez): The technology and so when we brought this to to the G seven during the Kenyan Presidents in 2018 there was a lot of pushback there was a lot of pushback from industry, there was a lot of pushback from some countries within the G seven to actually discuss this issue as an ethical issue.
44:37
MS Canada (Philippe-Andr? Rodriguez): and actually it ended up leading to what is the Canada France statement or artificial intelligence which is really the basis for what is now known as the global partnership on artificial intelligence. MS Canada (Philippe-Andr? Rodriguez): Which IBM is a is a very active member and Francesca obviously which is really one of the organizations thinking more practically rough, how do we turn these principles that we're seeing.
45:04
MS Canada (Philippe-Andr? Rodriguez): all over the place, and obviously totally agree JESSICA pieces absolutely foundational and we tried to have not things that I think everybody kind of was doing the mappings But then when. MS Canada (Philippe-Andr? Rodriguez): That these came up we kind of just said Look, we have, we have the authoritative fees, we don't need you know to do any of this work so that was great.
45:22
MS Canada (Philippe-Andr? Rodriguez): And, and so yeah global partnership on the I was was really one of these initiatives and really an initiative that seeks to be inclusive. MS Canada (Philippe-Andr? Rodriguez): Bringing you know, countries such as Brazil, India into the mix, a lot of global South outreach increasingly to to really have something that is more inclusive than as we've kind of noted before.
45:45
MS Canada (Philippe-Andr? Rodriguez): You know, a group of European and North American countries dictating what the rules of the road should look like. MS Canada (Philippe-Andr? Rodriguez): However, and as you know, I think, going back again to JESSICA, is not being a lot of this work has been done in the pace of web two three years.
45:59
MS Canada (Philippe-Andr? Rodriguez): And that this has been way too quick for a lot of countries and so when we see a lot of these conversations and even this weekend UNESCO we realized that there's there's still a huge gap. MS Canada (Philippe-Andr? Rodriguez): In terms of understanding of this epic space between countries that have had the sustained conversation for a while, such as Canada.
46:19
MS Canada (Philippe-Andr? Rodriguez): and a host of European countries versus some countries that might not have the capacity to literacy to engage in those normative conversations at the global level, and have a really solid. MS Canada (Philippe-Andr? Rodriguez): You know understanding of the issues and so that's been a bit of a challenge to really be inclusive, understand that a lot of work is is ongoing a really rapid base multiplication of.
46:44
MS Canada (Philippe-Andr? Rodriguez): The processes, but also being inclusive, in a sense of being able to bring along some of these groups that may not have the capacity to meaningfully engage. MS Canada (Philippe-Andr? Rodriguez): From illiteracy for perspective from a capacity perspective, for instance. MS Canada (Philippe-Andr? Rodriguez): So navigating that that has been quite difficult trying to remain inclusive but also keep up and continue to advance this this file.
47:08
MS Canada (Philippe-Andr? Rodriguez): So so that's really at the global level, and we, and we can go back we've talked about some of the challenges with the human rights language and why it's useful where it's not useful and we can come back to that, but. MS Canada (Philippe-Andr? Rodriguez): Just at the domestic level we've already mentioned Canada has a directive it's not a regulation but that's really technical.
47:29
MS Canada (Philippe-Andr? Rodriguez): On the use of Ai in government and really we strive to basically implement all of the principles that we've heard from from from previous speakers so transparency open by design. MS Canada (Philippe-Andr? Rodriguez): Human Rights framed and really.
47:47
MS Canada (Philippe-Andr? Rodriguez): it's, one that is constantly evolving so every six months it is being updated, because we know that technology is evolving and new challenges, new harms are brought to the floor, that we need to take into consideration, it is also risk based. MS Canada (Philippe-Andr? Rodriguez): because, obviously, the risk is not the same if you're talking about you know, using Ai to an Ai system to grant a visa versus using any Ai system to.
48:10
MS Canada (Philippe-Andr? Rodriguez): detect fish patterns in the ocean, for instance, right, so the risk is not necessarily the same and we need different. MS Canada (Philippe-Andr? Rodriguez): types of mechanisms to be able to address these different situations, Ai is is, we like to call it ai but really it's in the application that needs to think about the ethics, much more than in the technology itself as Francesca in particular said earlier.
48:32
MS Canada (Philippe-Andr? Rodriguez): we've also really focused on literacy, because we feel like in order to make the right. MS Canada (Philippe-Andr? Rodriguez): decisions, we need an informed citizenry that can participate in those conversations and this takes time. MS Canada (Philippe-Andr? Rodriguez): And this is partly why, in terms of regulation, I think, Canada has been a bit slower to have a general data protection regulation.
48:51
MS Canada (Philippe-Andr? Rodriguez): Because there was this feeling that we still need to do a lot of this ability literacy piece, so that decisions that are made are actually democratic and then can be made accountable and that there can be an accountability mechanism in in the public to actually ensure.
49:08
MS Canada (Philippe-Andr? Rodriguez): That you know we are going in the right direction ethically and in terms of you know, being entirely democratic and then, finally, yes, so going back to human rights, I think this is really a priority for us.
49:20
MS Canada (Philippe-Andr? Rodriguez): Why because it's already this baseline that we have right like there are definitely issues with human rights and it's definitely not enough, they are gaps within the human rights framework in terms of Ai ethics, I think. MS Canada (Philippe-Andr? Rodriguez): We all recognize that, for instance, environmental implications are completely excluded essentially from from this.
49:40
MS Canada (Philippe-Andr? Rodriguez): Some of the intersection of elements related to to to harms are difficult to actually grasp, in the context of human rights. MS Canada (Philippe-Andr? Rodriguez): Obviously responsibility of industry is sometimes difficult and especially in terms of enforcement mechanisms, despite the fact that we are, for instance, the UN guiding principles on business and human rights.
49:59
MS Canada (Philippe-Andr? Rodriguez): You know, we still have a lot of gaps and we still a ways to go. MS Canada (Philippe-Andr? Rodriguez): But I think, by and large, there is this this increasing recognition, even at UNESCO, I must say, even in the global side there is this recognition that starting from human rights and then building on top of that.
50:15
MS Canada (Philippe-Andr? Rodriguez): A more robust ethical framework is the way to go to start from somewhere where you know we all have this baseline we all, you know have international legal. MS Canada (Philippe-Andr? Rodriguez): obligations, this is countries that we can build from and then build something bigger so i'll leave it at that and definitely welcome to conversation on that Thank you. Elizabeth Renieris: Thank you so much, really, really appreciate your perspectives on this.
50:38
Elizabeth Renieris: As we are running short on time and we saw how starkly underrepresented, Latin America has been in this conversation. Elizabeth Renieris: I want to take moderators prerogative here to give fibro the rest of the time or as much as you need to give us your perspective on what are we missing in this conversation and how is the perspective different from where you are in Brazil. Fabro Steibel: And thank you very much. Fabro Steibel: credit for the lunch and then.
51:05
Fabro Steibel: How to follow the previous talks because it's very good so i'll bring two topics here I don't think they have appeared yet, although they. Fabro Steibel: They are transitioning to the others, and this is a perspective from Brazil, so in 2018 we did an event in huge and Adam for three days 40 countries 400 people on Ai ethics and inclusion they do so to find.
51:27
Fabro Steibel: To to identify an agenda for research and investments, mostly called away the most important finding for me is the survey, we did before we start event. Fabro Steibel: So we sent out a form saying what is one of the questions was what do you consider to be most important in the eye and ethics, including the agenda. Fabro Steibel: And then, if break down the results from the global north and the global South, the results are very different.
51:51
Fabro Steibel: From the global North it's about principle, this about implementation it's about the ethics of political South was about inclusion was about education. Fabro Steibel: And this is very important, as we know, to four years later, how the different in makes when you have less people talking when have less funds. Fabro Steibel: And then, it comes to the second point, which is polarization.
52:16
Fabro Steibel: When we didn't receive the Microsoft Internet the Brazilian Internet view of rice, this is a new stakeholder debate that we host. Fabro Steibel: In the Internet through the Internet with everyone okay you're not hold Brazil on the whole country but. Fabro Steibel: Even so, as a different way to the policy and the results are quite good Max do seven years old, we have over then and polity safe harbors a very interesting debate for digital right could we do that today.
52:43
Fabro Steibel: really hard to answer because polarization nowadays is just so strong so hard, and how does connect to access. Fabro Steibel: Funds for civil society funds for academia has gone down and fund for government and private sector has been stable, because they have other sources of funding. Fabro Steibel: And these increase it again, of how we can have a mood seat holder debate around Ai and ethics.
53:07
Fabro Steibel: i'm not complaining i'm just saying it's a different spectrum, so we learned a lesson on one SEC hold the way we love it but it's harder now to do a mood stakeholder debate and with the whole polarization. Fabro Steibel: You spend every two arguments one fighting is no five g's not about cancer, so you spend a lot of resources on fighting polarization. Fabro Steibel: And that leads me to the last debate on the global South perspective, which is the balance in between us and abuse.
53:39
Fabro Steibel: there's lots of discuss on the abuse of Ai ethics in a very important facial recognition for certainly something that we can easily talk about right. Fabro Steibel: You have bias on gender, you have buyers on race, are you have made us for security in a quite a silencer one. Fabro Steibel: At the same times, we do have the agenda of us use of facial recognition for digital identity can be inclusive, in a country where 30% doesn't have Internet in 40% doesn't have a bank account.
54:07
Fabro Steibel: The use of face recognition for every people to access lawyers and some public services. Fabro Steibel: it's very inclusive So how do you find a way to divide us and abuse without just focus on abuse and also using the US, and this is really hard so last week, for example, one of the main favelas Islam society organizations. Fabro Steibel: They wanted to very transparent and how food is distributed to sitters.
54:34
Fabro Steibel: And they decided to why not us based organization so i'm giving you rice, you are receiving race and then we faced with conditions we have transparency, the system. Fabro Steibel: And then all the questions cows wait, where is the data go into what have you go through the vulnerable groups, what what what. Fabro Steibel: So there is a important generally use, there is an important generally abuse and how do you combine this in the case of the.
54:58
Fabro Steibel: Implementation Islams, this is a bad they haven't they haven't seen the issues and the stock bed for 24 hours later. Fabro Steibel: But when you think about government implementing technology and so on, sometimes you stop ideas before they even start, and it can promote inclusion So how do we.
55:16
Fabro Steibel: discuss ethics, but without stopping the implementation and these can proceed of suicides are very hard sentence to say because usually we don't say hey you forgot that hey What about this hey minorities, and so on. Fabro Steibel: But there is an important agenda between us and abuse here, and this Palace and i'll close say what I learned so far on that you're trying to debate, which was 10 years ago that we had in Brazil.
55:40
Fabro Steibel: So it's very hard to to disagree and atrocities, you understand the Internet infrastructures, now we want good Internet for all but then, when you have a control 40% is connected. Fabro Steibel: You have to make a choice, do you want Internet now for all but not that good for for few or do you want to wait longer and have good Internet, and this is a very hard question.
56:02
Fabro Steibel: Because in 2021 30% of residents disconnected So how can I wait at the same times, if I have an Internet there's a garden, with just Facebook, is it better than it was before. Fabro Steibel: So the ethics principles for me it's a it's a place where you have to discuss the use and abuse.
56:23
Fabro Steibel: Considerably be the polarization is very costly and very hard to be moved stakeholder and where access plays a big deficit and we see this in the cyber security debate so Brazil have been have been victim of several. Fabro Steibel: Major cyber security events, one of the Supreme Court was targeted for one week of functioning. Fabro Steibel: Just like that we had more liquid data than citizens in Brazil and we are 200 millions, and so one So how do we do we balance so.
56:53
Fabro Steibel: Sometimes it's connected and what what others are saying, but coming from a resume perspective and the global South I know it's a different agenda, I noticed the same problem in I know we are connected and we are learning on the go. Elizabeth Renieris: Thank you so much for bro. Fabro Steibel: um. Elizabeth Renieris: This is exactly why we wanted you to be part of this conversation, because I think it's really important perspective that gets really lost.
57:19
Elizabeth Renieris: In the broader conversation sometimes and and you're so right to point out how different the interest in equities on the ground can be. Elizabeth Renieris: So, really, really appreciate you, adding that layer to this conversation we are really short on time here, but I think that we've had a really rich conversation, as it is so, unfortunately we won't be able to turn to any of the pre submitted questions.
57:44
Elizabeth Renieris: If anyone has any burning reactions or comments that they want to make in 30 seconds please feel free to jump in otherwise I think we're probably best to to wrap this up so. Elizabeth Renieris: Thank you so much again to just john Felipe fog bro and Francesca we hope to do this again, perhaps next year to revisit the state of Ai ethics and technology ethics more generally.
58:08
Elizabeth Renieris: And of course we'll stay in touch and share your work and your materials through the lab and thank you to everyone who's joined us today, you can follow us on Twitter at tech athletics lab. Elizabeth Renieris: or check out our brand new website at tech ethics nd.edu and we'll be sharing this video as well, so thanks again and everyone have a pleasant day.
58:28
Francesca Rossi -- IBM: Thank you Elizabeth for organizing these and for leading the Notre dam IBM ethics tech activists lab I think it's the promising very promising for not just the conversation like this one, but also practical, concrete actions to advance the state of Ai ethics thank. Elizabeth Renieris: Thank you so much, thanks everyone. Jessica Fjeld: thanks for having us. John Havens: Thanks everyone thanks Elizabeth. hi.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat