Stipends and Benefits • About the Berkman Klein Center • FAQ
Required Application Materials • Apply!
The Berkman Klein Center for Internet & Society at Harvard University is now accepting fellowship applications for the 2018-2019 academic year through our annual open call. This opportunity is for those who wish to spend 2018-2019 in residence in Cambridge, MA as part of the Center's vibrant community of research and practice, and who seek to engage in collaborative, cross-disciplinary, and cross-sectoral exploration of some of the Internet's most important and compelling issues.
Applications will be accepted until Wednesday, January 31, 2018 at 11:59 p.m. Eastern Time.
We invite applications from people working on a broad range of opportunities and challenges related to Internet and society, which may overlap with ongoing work at the Berkman Klein Center and may expose our community to new opportunities and approaches. We encourage applications from scholars, practitioners, innovators, engineers, artists, and others committed to understanding and advancing the public interest who come from -- and have interest in -- countries industrialized or developing, with ideas, projects, or activities in all phases on a spectrum from incubation to reflection.
About the Berkman Klein Fellowship Program
Through this annual open call, we seek to advance our collective work and give it new direction, and to deepen and broaden our networked community across backgrounds, disciplines, cultures, and home bases. We welcome you to read more about the program below, and to consider joining us as a fellow!
“The Berkman Klein Center's mission is to explore and understand cyberspace; to study its development, dynamics, norms, and standards; and to assess the need or lack thereof for laws and sanctions.
We are a research center, premised on the observation that what we seek to learn is not already recorded. Our method is to build out into cyberspace, record data as we go, self-study, and share. Our mode is entrepreneurial nonprofit.”
Inspired by our mission statement, the Berkman Klein Center’s fellowship program provides an opportunity for some of the world’s most innovative thinkers and changemakers to come together to hone and share ideas, find camaraderie, and spawn new initiatives. The program encourages and supports fellows in an inviting and playful intellectual environment, with community activities designed to foster inquiry and risk-taking, to identify and expose common threads across fellows’ individual activities, and to bring fellows into conversation with the faculty directors, employees, and broader community at the Berkman Klein Center. From their diverse backgrounds and wide-ranging physical and virtual travels, Berkman Klein Center fellows bring fresh ideas, skills, passion, and connections to the Center and our community, and from their time spent in Cambridge help build and extend new perspectives and actions out into the world.
A non-traditional appointment that defies any one-size-fits-all description, each Berkman Klein fellowship carries a unique set of opportunities, responsibilities, and expectations based on each fellow’s goals. Fellows appointed through this open call come into their fellowship with a personal research agenda and set of ambitions they wish to conduct while at the Center. These might include focused study or writing projects, action-oriented meetings, the development of a set of technical tools, capacity building efforts, testing different pedagogical approaches, or efforts to intervene in public discourse and trialing new platforms for exchange. Over the course of the year fellows advance their research and contribute to the intellectual life of the Center and fellowship program activities; as they learn with and are influenced by their peers, fellows have the freedom to change and modify their plans.
Together fellows actively design and participate in weekly all-fellows sessions, working groups, skill shares, hacking and development sessions, and shared meals, as well as joining in a wide-range of Berkman Klein Center events, classes, brainstorms, interactions, and projects. While engaging in both substance and process, much of what makes the fellowship program rewarding is created each year by the fellows themselves to address their own interests and priorities. These entrepreneurial, collaborative ventures – ranging at once from goal-oriented to experimental, from rigorous to humorous – ensure the dynamism of a fellowship experience, the fellowship program, and the Berkman Klein community. As well, the Center works to support our exemplary alumni network, and beyond a period of formal affiliation, community members maintain ongoing active communication and mutual support across cohorts.
back to top
Alongside and in conversation with the breadth and depth of topics explored through the Center’s research projects, fellows engage the fairly limitless expanse of Internet & society issues. Within each cohort of fellows we encourage and strive for wide inquisition and focused study, and these areas of speciality and exploration vary from fellow to fellow and year to year. Some broad issues of interest include (but are not limited to) fairness and justice; economic growth and opportunity; the ethics and governance of artificial intelligence; equity, agency, inclusion, and diversity; health; security; privacy; access to information; regulation; politics; and democracy. As fields of Internet and society studies continue to grow and evolve, and as the Internet reaches into new arenas, we expect that new areas of interest will emerge across the Center as well. We look forward to hearing from potential fellows in these nascent specialities and learning more about the impact of their work.
We welcome applications from people who feel that a year in our community as a fellow would accelerate their efforts and contribute to their ongoing personal and professional development.
Fellows come from across the disciplinary spectrum and different life paths. Some fellows are academics, whether students, post-docs, or professors. Others come from outside academia, and are technologists, entrepreneurs, lawyers, policymakers, activists, journalists, educators, or other types of practitioners from various sectors. Many fellows wear multiple hats, and straddle different pursuits at the intersections of their capacities. Fellows might be starting, rebooting, driving forward in, questioning, or pivoting from their established careers. Fellows are committed to spending their fellowship in concert with others guided by a heap of kindness, a critical eye, and a generosity of spirit.
The fellowship selection process is a multi-dimensional mix of art and science, based on considerations that are specific to each applicant and that also consider the composition of the full fellowship class. Please visit our FAQ to learn more about our selection criteria and considerations.
To learn more about the backgrounds of our current community of fellows, check out our fall video series with new fellows, 2017-2018 community announcement, read their bios, and find them on Twitter. As well, other previous fellows announcements give an overview of the people and topics in our community: 2016-2017, 2015-2016, 2014-2015, 2013-2014.
The work and well-being of the Berkman Klein Center for Internet & Society are profoundly strengthened by the diversity of our network and our differences in background, culture, experience, national origin, religion, sexual orientation, gender, gender identity, race, ethnicity, age, ability, and much more. We actively seek and welcome people of color, women, the LGBTQIA+ community, persons with disabilities, and people at intersections of these identities, from across the spectrum of disciplines and methods. In support of these efforts, we are offering a small number of stipends to select incoming fellows chosen through our open call for applications. More information about the available stipends may be found here. More information about the Center’s approach to diversity and inclusion may be found here.
While we embrace our many virtual connections, spending time together in person remains essential. In order to maximize engagement with the community, fellows are encouraged to spend as much time at the Center as they are able, and are expected to conduct much of their work from the Cambridge area, in most cases requiring residency. Tuesdays hold particular importance--it is the day the fellows community meets for a weekly fellows hour, as well as the day the Center hosts a public luncheon series; as a baseline we ask fellows to commit to spending as many Tuesdays at the Center as possible.
Fellowship terms run for one year, and we generally expect active participation from our fellows over the course of the academic year, roughly from the beginning of September through the end of May.
In some instances, fellows are re-appointed for consecutive fellowship terms or assume other ongoing affiliations at the Center after their fellowship.
Berkman Klein fellowships awarded through the open call for applications are rarely stipended, and most fellows receive no direct funding through the Berkman Klein Center as part of their fellowship appointment.
To make Berkman Klein fellowships a possibility for as wide a range of applicants as possible, in the 2018-2019 academic year we will award a small number of stipends to select incoming fellows chosen through our open call for applications. This funding is intended to support people from communities who are underrepresented in fields related to Internet and society, who will contribute to the diversity of the Berkman Klein Center’s research and activities, and who have financial need. More information about this funding opportunity can be found here.
There are various ways fellows selected through the open call might be financially supported during their fellowship year. A non-exhaustive list: some fellows have received external grants or awards in support of their research; some fellows have received a scholarship or are on sabbatical from a home institution; some fellows do consulting work; some fellows maintain their primary employment alongside their fellowship. In each of these different scenarios, fellows and the people with whom they work have come to agreements that allow the fellow to spend time and mindshare with the Berkman Klein community, with the aim to have the fellow and the work they will carry out benefit from the affiliation with the Center and the energy spent in the community. Fellows are expected to independently set these arrangements with the relevant parties.
Office and Meeting Space
We endeavor to provide comfortable and productive spaces for for coworking and flexible use by the community. Some Berkman Klein fellows spend every day in our office, and some come in and out throughout the week while otherwise working from other sites. Additionally, fellows are supported in their efforts to host small meetings and gatherings at the Center and in space on the Harvard campus.
Access to University Resources
Library Access: Fellows are able to acquire Special Borrower privileges with the Harvard College Libraries, and are granted physical access into Langdell Library (the Harvard Law School Library). Access to the e-resources is available within the libraries.
Courses: Berkman Klein fellows often audit classes across Harvard University, however must individually ask for permission directly from the professor of the desired class.
Benefits: Fellows appointed through the open call do not have the ability to purchase University health insurance or get Harvard housing.
The Berkman Klein Center for Internet & Society at Harvard University is dedicated to exploring, understanding, and shaping the development of the digitally-networked environment. A diverse, interdisciplinary community of scholars, practitioners, technologists, policy experts, and advocates, we seek to tackle the most important challenges of the digital age while keeping a focus on tangible real-world impact in the public interest. Our faculty, fellows, staff and affiliates conduct research, build tools and platforms, educate others, form bridges and facilitate dialogue across and among diverse communities. More information at https://cyber.harvard.edu.
To learn more about the Center’s current research, consider watching a video of the Berkman Klein Center’s Faculty Chair Jonathan Zittrain giving a lunch talk from Fall 2017, and check out the Center’s most recent annual reports.
back to top
To hear more from former fellows, check out 15 Lessons from the Berkman Fellows Program, a report written by former fellow and current Fellows Advisory Board member David Weinberger. The report strives to "explore what makes the Berkman Fellows program successful...We approached writing this report as a journalistic task, interviewing a cross-section of fellows, faculty, and staff, including during a group session at a Berkman Fellows Hour. From these interviews a remarkably consistent set of themes emerged."
More information about fellows selection and the application process can be found on our Fellows Program FAQ.
If you have questions not addressed in the FAQ, please feel welcome to reach out Rebecca Tabasky, the Berkman Klein Center's manager of community programs, at firstname.lastname@example.org.
back to top
(1.) A current resume or C.V.
(2.) A personal statement that responds to the following two questions. Each response should be between 250-500 words.
What is the research you propose to conduct during a fellowship year? Please
describe the problems are you trying to solve;
outline the methods which might inform your research; and
tell us about the public interest and/or the communities you aim to serve through your work.
Why is the Berkman Klein Center the right place for you to do this work? Please share thoughts on:
how the opportunity to engage colleagues from different backgrounds -- with a range of experiences and training in disciplines unfamiliar to you -- might stimulate your work;
which perspectives you might seek out to help you fill in underdeveloped areas of your research;
what kinds of topics and skills you seek to learn with the Center that are outside of your primary research focus and expertise; and
the skills, connections, and insights you are uniquely suited to contribute to the Center’s community and activities.
(3.) A copy of a recent publication or an example of relevant work. For a written document, for instance, it should be on the order of a paper or chapter - not an entire book or dissertation - and should be in English.
(4.) Two letters of recommendation, sent directly from the reference.
back to top
The application deadline is Wednesday, January 31, 2018 at 11:59 p.m. Eastern Time.
Applications will be submitted online through our Application Tracker tool at:
Applicants will submit their resume/C.V., their personal statement, and their work sample as uploads within the Berkman Klein Application Tracker. Applicants should ensure that their names are included on each page of their application materials.
Recommendation letters will be captured through the Application Tracker, and the Application Tracker requires applicants to submit the names and contact information for references in advance of the application deadline. References will receive a link at which they can upload their letters. We recommend that applicants create their profiles and submit reference information in the Application Tracker as soon as they know they are going to apply and have identified their references - this step will not require other fellowship application materials to be submitted at that time. We do ask that letters be received from the references by the application deadline.
Instructions for creating an account and submitting an application through the Application Tracker may be found here.
back to top
What happens when our criminal justice system uses algorithms to help judges determine bail, sentencing, and parole?Thumbnail Image:
Earlier this month, a group of researchers from Harvard and MIT directed an open letter to the Massachusetts Legislature to inform its consideration of risk assessment tools as part of ongoing criminal justice reform efforts in the Commonwealth. Risk assessment tools are pieces of software that courts use to assess the risk posed by a particular criminal defendant in a particular set of circumstances. Senate Bill 2185 — passed by the Massachusetts Senate on October 27, 2017 — mandates implementation of RA tools in the pretrial stage of criminal proceedings.
In this episode of the Berkman Klein Center podcast, The Platform, Managing Director of the Cyberlaw Clinic Professor Chris Bavitz discusses some of the concerns and opportunities related to the use of risk assessment tools as well as some of the related work the Berkman Klein Center is doing as part of the Ethics and Governance of AI initiative in partnership with the MIT Media Lab.
What need are risk assessment tools addressing? Why would we want to implement them?
Well, some people would say that they’re not addressing any need and ask why we would ever use a computer program when doing any assessments. But I think that there are some ways in which they’re helping to solve problems, particularly around consistency. Another potential piece of it, and this is where we start to get sort of controversial, is that the criminal justice system is very biased and has historically treated racial minorities and other members of marginalized groups poorly. A lot of that may stem from human biases that creep in anytime you have one human evaluating another human being. So there’s an argument to be made that if we can do risk scoring right and turn it into a relatively objective process, we might remove from judges the kind of discretion that leads to biased decisions.
Are we there yet? Can these tools eliminate bias like that?
My sense is that from a computer science perspective we’re not there. In general, these kinds of technologies that use machine learning are only as good as the data on which they’re trained. So if I’m trying to decide whether you’re going to come back for your hearing in six months, the only information that I have to train a risk scoring tool to give me a good prediction on that front is data about people like you who came through the criminal justice system in the past. And if we take as a given that the whole system is biased, then the data is that coming out of that system is biased. And when we feed that data to a computer program, the results are going to be biased.
And we don’t know what actually goes into these tools?
Many of the tools that are in use in states around the country are tools that are developed by private companies. So with most of the tools we do not have a very detailed breakdown of what factors are being considered, what relative weights are being given to each factor, that sort of thing. So one of the pushes for advocates in this area is that at the very least we need more transparency.
Tell me about the Open Letter to the Legislature. Why did you write it?
The Massachusetts Senate and House are in the process of considering criminal justice reform broadly speaking in Massachusetts. The Senate bill has some language in it that suggests that risk scoring tools should be adopted in the Commonwealth and that we should take steps to make sure that they’re not biased. And a number of us, most of whom are involved in the Berkman and MIT Media Lab AI Ethics and Governance efforts, signed onto this open letter to the Mass Legislature that basically said, “Look these kinds of tools may have a place in the system, but simply saying ‘Make sure they’re not biased’ is not enough. And if you’re going to go forward, here are a whole bunch of principles that we want you to adhere to,” basically trying to set up processes around both the procurement or development of the tool, the implementation of the tool, the training of the judges on how to use it and what the scores really mean and how they should fit into their legal analysis, and then ultimately the rigorous evaluation of the outcomes. Are these tools actually having the predictive value that was promised? How are we doing on the bias front? Does this seem to be generating results that are biased in statistically significant ways?
What are you hoping will happen next?
I think we would view part of our mission here at Berkman Klein as making sure that this is the subject of vigorous debate. Informed debate, to be clear, because I think that sometimes the debate about this devolves into either that technology is going to solve all our problems, or it’s a dystopian future with robotic judges that are going to sentence us to death, and I don’t think it’s either of those things. Having this conversation in a way that is nuanced and responsible will be really difficult, but I think it’s something we absolutely have to do.
This initiative at Berkman Klein and MIT is the Ethics and Governance of Artificial Intelligence Initiative, but there’s nothing about anything we’ve talked about here that really has to do with artificial intelligence where the computer program is learning and evolving and changing and adapting over time. But that’s coming. And the more we get used to these kinds of systems working in the criminal justice system and spitting out risk scores that judges take into account, the more comfortable we’re going to be as the computing power increases and the autonomy of these programs increases.
I don’t mean to be too dystopic about it and say that bad stuff is coming, but it’s only a matter of time. It’s happening in our cars, and it’s happening in our news feeds on social media sites. More and more decisions are being made by algorithms. And anytime we get a technological intervention in a system like this, particularly where people’s freedom is at stake, I think we want to tread really carefully, recognizing that the next iteration of this technology is going to be more extensive, and raise even more challenging questions.
Tuesday, December 12, 2017 at 12:00 pm
Berkman Klein Center for Internet & Society at Harvard University
Harvard Law School campus
Wasserstein Hall, Milstein East C (HLS campus map)
RSVP required to attend in person
Event will be live webcast at 12:00 pm
Since the rise of the web in the 1990s, technological skeptics have always faced resistance. To question the virtue and righteousness of tech, and especially computing, was seen as truculence, ignorance, or luddism. But today, the real downsides of tech, from fake news to data breaches to AI-operated courtrooms to energy-sucking bitcoin mines, have become both undeniable and somewhat obvious in retrospect.
In light of this new technological realism, perhaps there is appetite for new ways to think about and plan for the future of technology, which anticipates what might go right and wrong once unproven tech mainstreams quickly. As a test case, this talk will consider a technology that has not yet mainstreamed—autonomous vehicles—as a test case.
Dr. Ian Bogost is an author and an award-winning game designer. He is Ivan Allen College Distinguished Chair in Media Studies and Professor of Interactive Computing at the Georgia Institute of Technology, where he also holds an appointment in the Scheller College of Business. Bogost is also Founding Partner at Persuasive Games LLC, an independent game studio, and a Contributing Editor at The Atlantic. He is the author or co-author of ten books including Unit Operations: An Approach to Videogame Criticism and Persuasive Games: The Expressive Power of Videogames.
Bogost’s videogames about social and political issues cover topics as varied as airport security, consumer debt, disaffected workers, the petroleum industry, suburban errands, pandemic flu, and tort reform. His games have been played by millions of people and exhibited or held in collections internationally, at venues including the Smithsonian American Art Museum, the Telfair Museum of Art, The San Francisco Museum of Modern Art, The Museum of Contemporary Art, Jacksonville, the Laboral Centro de Arte, and The Australian Centre for the Moving Image.
His independent games include Cow Clicker, a Facebook game send-up of Facebook games that was the subject of a Wired magazine feature, and A Slow Year, a collection of videogame poems for Atari VCS, Windows, and Mac, which won the Vanguard and Virtuoso awards at the 2010 IndieCade Festival.
Bogost holds a Bachelors degree in Philosophy and Comparative Literature from the University of Southern California, and a Masters and Ph.D. in Comparative Literature from UCLA. He lives in Atlanta.
Jeffrey is Professor of Romance Languages & Literature, Harvard Graduate School of Design; Director, metaLAB (at) Harvard; and Director, Berkman Klein Center for Internet & Society. A cultural historian with research interests extending from Roman antiquity to the present, his most recent books are The Electric Information Age Book (a collaboration with the designer Adam Michaels (Princeton Architectural Press, 2012) and Italiamerica II (Il Saggiatore, 2012). His pioneering work in the domains of digital humanities and digitally augmented approaches to cultural programming includes curatorial collaborations with the Triennale di Milano, the Cantor Center for the Visual Arts, the Wolfsonian-FIU, and the Canadian Center for Architecture. His Trento Tunnels project — a 6000 sq. meter pair of highway tunnels in Northern Italy repurposed as a history museum– was featured in the Italian pavilion of the 2010 Venice Biennale and at the MAXXI in Rome in RE-CYCLE - Strategie per la casa la città e il pianeta (fall-winter 2011). He is Professor of Romance Languages & Literature, on the teaching faculty of Harvard’s Graduate School of Design,and is the faculty director of metaLAB (at) Harvard.
Links to selected writing over the last year or so that are relevant:
An international symposium aimed at building capacity and exploring ideas for data democratization and inclusion in the age of AI.Thumbnail Image:
AI-based technologies — and the vast datasets that power them — are reshaping a broad range of sectors of the economy and are increasingly affecting the ways in which we live our lives. But to date these systems remain largely the province of a few large companies and powerful nations, raising concerns over how they might exacerbate inequalities and perpetuate bias against underserved and underrepresented populations.
In early November, on behalf of a global group of Internet research centers known as the Global Network of Internet & Society Centers (NoC) , the Institute for Technology & Society of Rio de Janeiro and the Berkman Klein Center for Internet & Society at Harvard University co-organized a three-day symposium on these topics in Brazil. The event brought together representatives from academia, advocacy groups, philanthropies, media, policy, and industry from more than 20 nations to start identifying and implementing ways to make the class of technologies broadly termed “AI” more inclusive.
The symposium — attended by about 170 people from countries including Nigeria, Uganda, South Africa, Kenya, Egypt, India, Japan, Turkey, and numerous Latin American and European nations — was intended to build collaborative partnerships and identify research questions as well as action items. These may include efforts to draft a human rights or regulatory framework for AI; define ways to democratize data access and audit algorithms and review their effects; and commit to designing and deploying AI that incorporates the perspectives of traditionally underserved and underrepresented groups, which include urban and rural poor communities, women, youth, LGBTQ individuals, ethnic and racial groups, and people with disabilities.
AI-based systems are “black boxes,” resulting in massive information asymmetries between the developers of such systems and consumers and policymakers. In order to bridge this information gap, this article proposes a conceptual framework for thinking about governance for AI.Publication Date 20 Nov 2017 Author(s) Download from DASHDownload from IEEE Internet Computing
AI-based systems are “black boxes,” resulting in massive information asymmetries between the developers of such systems and consumers and policymakers. In order to bridge this information gap, this article proposes a conceptual framework for thinking about governance for AI.
Many sectors of society rapidly adopt digital technologies and big data, resulting in the quiet and often seamless integration of AI, autonomous systems, and algorithmic decision-making into billions of human lives. AI and algorithmic systems already guide a vast array of decisions in both private and public sectors. For example, private global platforms, such as Google and Facebook, use AIbased filtering algorithms to control access to information. AI algorithms that control self-driving cars must decide on how to weigh the safety of passengers and pedestrians. Various applications, including security and safety decisionmaking systems, rely heavily on A-based face recognition algorithms. And a recent study from Stanford University describes an AI algorithm that can deduce the sexuality of people on a dating site with up to 91 percent accuracy. Voicing alarm at the capabilities of AI evidenced within this study, and as AI technologies move toward broader adoption, some voices in society have expressed concern about the unintended consequences and potential downsides of widespread use of these technologies.
To ensure transparency, accountability, and explainability for the AI ecosystem, our governments, civil society, the private sector, and academia must be at the table to discuss governance mechanisms that minimize the risks and possible downsides of AI and autonomous systems while harnessing the full potential of this technology. Yet the process of designing a governance ecosystem for AI, autonomous systems, and algorithms is complex for several reasons. As researchers at the University of Oxford point out,3 separate regulation solutions for decision-making algorithms, AI, and robotics could misinterpret legal and ethical challenges as unrelated, which is no longer accurate in today’s systems. Algorithms, hardware, software, and data are always part of AI and autonomous systems. To regulate ahead of time is dicult for any kind of industry. Although AI technologies are evolving rapidly, they are still in the development stages. A global AI governance system must be flexible enough to accommodate cultural dierences and bridge gaps across dierent national legal systems. While there are many approaches we can take to design a governance structure for AI, one option is to take inspiration from the development and evolution of governance structures that act on the Internet environment. Thus, here we discuss dierent issues associated with governance of AI systems, and introduce a conceptual framework for thinking about governance for AI, autonomous systems, and algorithmic decision-making processes.Producer Intro Authored by
The paper reviews current societal, moral, and legal norms around explanations, and then focuses on the different contexts under which an explanation is currently required under the law. It ultimately finds that, at least for now, AI systems can and should be held to a similar standard of explanation as humans currently are.Publication Date 27 Nov 2017 Thumbnail Image: External Links: Download from SSRNDownload from DASHDownload from arXiv.org
for the Berkman Klein Center Working Group on Explanation and the Law:
Chris Bavitz, Harvard Law School; Berkman Klein Center for Internet & Society at Harvard University
Ryan Budish, Berkman Klein Center for Internet & Society at Harvard University
Finale Doshi-Velez, John A. Paulson School of Engineering and Applied Sciences, Harvard University
Sam Gershman, Department of Psychology and Center for Brain Science, Harvard University
Mason Kortz, Harvard Law School Cyberlaw Clinic
David O'Brien, Berkman Klein Center for Internet & Society at Harvard University
Stuart Shieber, John A. Paulson School of Engineering and Applied Sciences, Harvard University
James Waldo, John A. Paulson School of Engineering and Applied Sciences, Harvard University
David Weinberger, Berkman Klein Center for Internet & Society at Harvard University
Alexandra Wood, Berkman Klein Center for Internet & Society at Harvard University
The ubiquity of systems using artificial intelligence or "AI" has brought increasing attention to how those systems should be regulated. The choice of how to regulate AI systems will require care. AI systems have the potential to synthesize large amounts of data, allowing for greater levels of personalization and precision than ever before|applications range from clinical decision support to autonomous driving and predictive policing. That said, common sense reasoning [McCarthy, 1960] remains one of the holy grails of AI, and there exist legitimate concerns about the intentional and unintentional negative consequences of AI systems [Bostrom, 2003, Amodei et al., 2016, Sculley et al., 2014].
There are many ways to hold AI systems accountable. In this work, we focus on one: explanation. Questions about a legal right to explanation from AI systems was recently debated in the EU General Data Protection Regulation [Goodman and Flaxman, 2016, Wachter et al., 2017], and thus thinking carefully about when and how explanation from AI systems might improve accountability is timely. Good choices about when to demand explanation can help prevent negative consequences from AI systems, while poor choices may not only fail to hold AI systems accountable but also hamper the development of much-needed beneficial AI systems.
Below, we briefly review current societal, moral, and legal norms around explanation, and then focus on the different contexts under which explanation is currently required under the law. We find that there exists great variation around when explanation is demanded, but there also exists important consistencies: when demanding explanation from humans, what we typically want to know is how and whether certain input factors affected the final decision or outcome.
These consistencies allow us to list the technical considerations that must be considered if we desired AI systems that could provide kinds of explanations that are currently required of humans under the law. Contrary to popular wisdom of AI systems as indecipherable black boxes, we find that this level of explanation should often be technically feasible but may sometimes be practically onerous|there are certain aspects of explanation that may be simple for humans to provide but challenging for AI systems, and vice versa. As an interdisciplinary team of legal scholars, computer scientists, and cognitive scientists, we recommend that for the present, AI systems can and should be held to a similar standard of explanation as humans currently are; in the future we may wish to hold an AI to a different standard.
The authors have invited researchers, technologists, and policy makers to engage with the ideas outlined in the paper by emailing email@example.com and firstname.lastname@example.org. For questions and comments related to broader AI themes and/or related activities of the Ethics and Governance of Artificial Intelligence Initiative, please email email@example.com.Producer Intro Authored by
As we integrate artificial intelligence deeper into our daily technologies, it becomes important to ask “why” not just of people, but of systems. A new working paper from the Berkman Klein Center at Harvard University and the MIT Media Lab maps out critical starting points for thinking about explanation in AI systems.Thumbnail Image:
“Why did you do that?” The right to ask that deceptively simple question and expect an answer creates a social dynamic of interpersonal accountability. Accountability, in turn, is the foundation of many important social institutions, from personal and professional trust to legal liability to governmental legitimacy and beyond.
As we integrate artificial intelligence deeper into our daily technologies, it becomes important to ask “why” not just of people, but of systems. However, human and artificial intelligences are not interchangeable. Designing an AI system to provide accurate, meaningful, human-readable explanations presents practical challenges, and our responses to those challenges may have far-reaching consequences. Setting guidelines for AI-generated explanations today will help us understand and manage increasingly complex systems in the future.
In response to these emerging questions, a new working paper from the Berkman Klein Center at Harvard University and the MIT Media Lab maps out critical starting points for thinking about explanation in AI systems. “Accountability of AI Under the Law: The Role of Explanation” is now available to scholars, policy makers, and the public.
“If we’re going to take advantage of all that AIs have to offer, we’re going to have to find ways to hold them accountable,” said Finale Doshi-Velez of Harvard’s John A. Paulson School of Engineering and Applied Sciences, “Explanation is one tool toward that end. We see a complex balance of costs and benefits, social norms, and more. To ground our discussion in concrete terms, we looked to ways that explanation currently functions in law.”
Doshi-Velez and Mason Kortz of the Berkman Klein Center and Harvard Law School Cyberlaw Clinic are lead authors of the paper, which is the product of an extensive collaboration within the Ethics and Governance of Artificial Intelligence Initiative, now underway at Harvard and MIT.
“An explanation, as we use the term in this paper, is a reason or justification for a specific decision made by an AI system--how a particular set of inputs lead to a particular outcome,” said Kortz. “A helpful explanation will tell you something about this process, such as the degree to which an input influenced the outcome, whether changing a certain factor would have changed the decision, or why two similar-looking cases turned out differently.”
The paper reviews current societal, moral, and legal norms around explanations, and then focuses on the different contexts under which an explanation is currently required under the law. It ultimately finds that, at least for now, AI systems can and should be held to a similar standard of explanation as humans currently are.
“It won’t necessarily be easy to produce explanations from complex AI systems that are processing enormous amounts of data,” Kortz added. “Humans are naturally able to describe our internal processes in terms of cause and effect, although not always with great accuracy. AIs, on the other hand, will have to be intentionally designed with the capacity to generate explanations in mind. This paper is the starting point for a series of discussions that will be increasingly important in the years ahead. We’re hoping this generates some constructive feedback from inside and outside the Initiative.”
Guided by the Berkman Klein Center at Harvard and the MIT Media Lab, the Ethics and Governance of Artificial Intelligence Initiative aims to foster global conversations among scholars, experts, advocates, and leaders from a range of industries. By developing a shared framework to address urgent questions surrounding AI, the Initiative aims to help public and private decision-makers understand and plan for the effective use of AI systems for the public good. More information at: https://cyber.harvard.edu/research/ai
This Comment, published in the JOLT Digest, is the first in a two-part series on how lawyers should think about art generated by artificial intelligences, particularly with regard to copyright law. This first part charts the anatomy of the AI-assisted artistic process. Thumbnail Image:
This Comment by Jessica Fjeld and Mason Kortz originally published in the Journal of Law and Technology's online Digest, is the first in a two-part series on how lawyers should think about art generated by artificial intelligences, particularly with regard to copyright law. This first part charts the anatomy of the AI-assisted artistic process. The second Comment in the series examine how copyright interests in these elements interact and provide practice tips for lawyers drafting license agreements or involved in disputes around AI-generated artwork.
Advanced algorithms that display cognition-like processes, popularly called artificial intelligences or “AIs,” are capable of generating sophisticated and provocative works of art. These technologies differ from widely-used digital creation and editing tools in that they are capable of developing complex decision-making processes, leading to unexpected outcomes. Generative AI systems and the artwork they produce raise mind-bending questions of ownership, from broad policy concerns to the individual interests of the artists, engineers, and researchers undertaking this work. Attorneys, too, are beginning to get involved, called on by their clients to draft licenses or manage disputes.
The Harvard Law School Cyberlaw Clinic at the Berkman Klein Center for Internet & Society has recently developed a practice in advising clients in the emerging field at the intersection of art and AI. We have seen for ourselves how attempts to negotiate licenses or settle disputes without a common understanding of the systems involved may result in vague and poorly understood agreements, and worse, unnecessary conflict between parties. More often than not, this friction arises between reasonable parties who are open to compromise, but suffer from a lack of clarity over what, exactly, is being negotiated. In the course of solving such problems, we have dissected generative AIs and studied their elements from a legal perspective. The result is an anatomy that forms the foundation of our thinking—and our practice—on the subject of AI-generated art. When the parties to an agreement or dispute share a common vocabulary and understanding of the nature of the work, many areas of potential conflict evaporate.
CopyrightX is a networked course that explores the current law of copyright; the impact of that law on art, entertainment, and industry; and the ongoing debates concerning how the law should be reformed.Thumbnail Image:
The application for the CopyrightX online sections will be open from Oct. 16 - Dec. 13. See CopyrightX:Sections for details.
CopyrightX is a networked course that explores the current law of copyright; the impact of that law on art, entertainment, and industry; and the ongoing debates concerning how the law should be reformed. Through a combination of recorded lectures, assigned readings, weekly seminars, and live interactive webcasts, participants in the course examine and assess the ways in which the copyright system seeks to stimulate and regulate creative expression.
In 2013, HarvardX, Harvard Law School, and the Berkman Klein Center for Internet & Society launched an experiment in distance education: CopyrightX, the first free and open distance learning course on law. After five successful offerings, CopyrightX is an experiment no longer. Under the leadership of Professor William Fisher, who created and directs the course, CopyrightX will be offered for a sixth time from January to May 2018.
Three types of courses make up the CopyrightX Community:
• a residential course on Copyright Law, taught by Prof. Fisher to approximately 100 Harvard Law School students;
• an online course divided into sections of 25 students, each section taught by a Harvard Teaching Fellow;
• a set of affiliated courses based at educational institutions worldwide, each taught by an expert in copyright law.
Participation in the 2018 online sections is free and is open to anyone at least 13 years of age, but enrollment is limited. Admission to the online sections will be administered through an open application process that ends on December 13, 2017. We welcome applicants from all countries, as well as lawyers and non-lawyers alike. To request an application, visit http://brk.mn/applycx18. For more details, see CopyrightX:Sections. (The criteria for admission to each of the affiliated courses are set by the course’s instructor. Students who will enroll in the affiliated courses may not apply to the online sections.)
We encourage widespread promotion of the application through personal and professional networks and social media. Feel free to circulate:
• this blog post
• the application page
An Open Letter to the Members of the Massachusetts Legislature Regarding the Adoption of Actuarial Risk Assessment Tools in the Criminal Justice System
This open letter — signed by Harvard and MIT-based faculty, staff, and researchers— is directed to the Massachusetts Legislature to inform its consideration of risk assessment tools as part of ongoing criminal justice reform efforts in the Commonwealth.Publication Date 9 Nov 2017 External Links: Download from DASHRead the letter on Medium
This open letter — signed by Harvard and MIT-based faculty, staff, and researchers Chelsea Barabas, Christopher Bavitz, Ryan Budish, Karthik Dinakar, Cynthia Dwork, Urs Gasser, Kira Hessekiel, Joichi Ito, Ronald L. Rivest, Madars Virza, and Jonathan Zittrain — is directed to the Massachusetts Legislature to inform its consideration of risk assessment tools as part of ongoing criminal justice reform efforts in the Commonwealth.
Producer Intro Authored by
Tuesday, November 28, 2017 at 12:00 pm
Berkman Klein Center for Internet & Society at Harvard University
Harvard Law School campus
Wasserstein Hall, Milstein East C, Room 2036 (HLS campus map)
RSVP required to attend in person
Watch Live Starting at 12pm
If you experience a video disruption reload to refresh the webcast.
We are pleased to welcome back Berkman Klein Fellow alumnus, Dennis Tenen, who joins us to discuss his new book, Plain Text: The Poetics of Computation (Stanford UP, 2017).
This book challenges the ways we read, write, store, and retrieve information in the digital age. Computers—from electronic books to smart phones—play an active role in our social lives. Our technological choices thus entail theoretical and political commitments. Dennis Tenen takes up today's strange enmeshing of humans, texts, and machines to argue that our most ingrained intuitions about texts are profoundly alienated from the physical contexts of their intellectual production. Drawing on a range of primary sources from both literary theory and software engineering, he makes a case for a more transparent practice of human–computer interaction. Plain Text is thus a rallying call, a frame of mind as much as a file format. It reminds us, ultimately, that our devices also encode specific modes of governance and control that must remain available to interpretation.
Dennis Tenen's research happens at the intersection of people, texts, and technology.
His recent work appears on the pages of Amodern, boundary 2, Computational Culture, Modernism/modernity, New Literary History, Public Books, and LA Review of Books on topics that range from book piracy to algorithmic composition, unintelligent design, and history of data visualization.
He teaches a variety of classes in fields of literary theory, new media studies, and critical computing in the humanities.
Tenen is a co-founder of Columbia's Group for Experimental Methods in the Humanities and author of Plain Text: The Poetics of Computation (Stanford UP, 2017).
For an updated list of projects, talks, and publications please visit dennistenen.com.
The Harvard Open Access Project (HOAP) at the Berkman Klein Center for Internet & Society is hiring a part-time research assistant!
The Harvard Open Access Project (HOAP) fosters open access to research, within Harvard and beyond, using a combination of education, consultation, collaboration, research, tool-building, and direct assistance. HOAP is a project within the Berkman Klein Center for Internet & Society at Harvard University. For more detail, see the project home page at http://cyber.harvard.edu/hoap.
The Research Assistant will contribute to the Open Access Tracking Project (OATP), using the TagTeam social-tagging platform, contribute to the Open Access Directory (OAD), and perform occasional research, help with grant reporting, and strategize about open access inside and outside Harvard University. The position offers remote work options, flexible scheduling, and community work spaces at the Berkman Klein Center for Internet & Society.
The position will remain open until the job is filled, and plan to begin reviewing applicants as soon as possible.
Work Requirements/Benefits Information:
This part-time position is 17.25 hours per week. The pay is at a rate of $11.50+ per hour, with the possibility of more to suit qualifications and experience. This position does not include benefits. The role will include the expectation of regular weekend work as needed to support time-sensitive projects (approximately 2 - 4 of total 17.25). The Research Assistant must be based in Massachusetts. The work may be done remotely, but will include regular face-to-face meetings in Cambridge, Massachusetts to review progress and discuss new ideas. Unfortunately we are not able to sponsor a visa for this position. This position is approved through the end of August, 2018.
Please send your current CV or resume and a cover letter summarizing your interest and experience to Peter Suber at firstname.lastname@example.org with “HOAP application” in the subject line.