布鲁金斯学会-人工智能实地合作：全球范围的人工智能研发（英）-2022.10-26正式版.pdf

《布鲁金斯学会-人工智能实地合作：全球范围的人工智能研发（英）-2022.10-26正式版.pdf》由会员分享，可在线阅读，更多相关《布鲁金斯学会-人工智能实地合作：全球范围的人工智能研发（英）-2022.10-26正式版.pdf（26页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、GLOBAL AI COOPERATION ON THE GROUNDAI RESEARCH AND DEVELOPMENT ON A GLOBAL SCALE REPORTOCTOBER 2022CAMERON F.KERRY|JOSHUA P.MELTZER|ANDREA RENDAFORUM FOR COOPERATION ON ARTIFICIAL INTELLIGENCEGLOBAL AI COOPERATION ON THE GROUND|AI RESEARCH AND DEVELOPMENT ON A GLOBAL SCALE Brookings Institution 1 In

2、troduction The Forum for Cooperation on Artificial Intelligence(FCAI)has investigated opportunities and obstacles for international cooperation to foster development of responsible artificial intelligence(AI).1 It has brought together officials from seven governments(Australia,Canada,the European Un

3、ion,Japan,Singapore,the United Kingdom,and United States with experts from industry,academia,and civil society to explore similarities and differences in national policies on AI,avenues of international cooperation,ecosystems of AI research and development(R&D),and AI standards development among oth

4、er issues.Following a series of roundtables in 2020 and 2021,we issued a progress report in October 2021 that articulated why international cooperation is especially needed on AI,identified significant challenges to such cooperation,and proposed four key areas where international cooperation could d

5、eepen:Regulatory alignment,standards development,trade agreements,and joint R&D.2 The report made 15 recommendations on ways to make progress in these areas.For joint R&D,recommendation R15 of the progress report called for development of“common criteria and governance arrangements for international

6、 large-scale AI R&D projects,”with the Human Genome Project(HGP)3 and the European Organization for Nuclear Research(CERN)4 as examples of the scale and ambition needed.The report summarized this recommendation as follows:“Joint research and development applying to large-scale global problems such a

7、s climate change or disease prevention and treatment can have two valuable effects:It can bring additional resources to the solution of pressing global challenges,and the collaboration can help to find common ground in addressing differences in approaches to AI.FCAI will seek to incubate a concrete

8、roadmap on such R&D for adoption by FCAI participants as well as other governments and international organizations.Using collaboration on R&D as a mechanism to work through matters that affect international cooperation on AI policy means that this recommendation should play out in the near term.”5 F

9、CAI convened a roundtable on February 10,2022,to explore specific use cases that may be candidates for joint international research and development and inform selection and design of such projects based on criteria outlined below.Potential areas considered were climate change,public health,privacy-e

10、nhancing technologies for sharing data,and improved tracking of economic growth and performance(economic measurement).This working paper distills the discussions and our analysis and research.We recommend that FCAI governments,stakeholders,and other likeminded entities prioritize cooperative R&D eff

11、orts on(1)deployment of AI as a tool for climate change monitoring and management and(2)accelerating development and adoption of privacy enhancing technologies(PETs).These distinctly different subject areas reflect the complex interaction of criteria explored in our discussions and appear to offer t

12、he most promising avenues of progress for international cooperation on AI R&D.Both areas emerged as clear favorites in discussion and polling among participants at the February 10 FCAI dialogue for the GLOBAL AI COOPERATION ON THE GROUND|AI RESEARCH AND DEVELOPMENT ON A GLOBAL SCALE Brookings Instit

13、ution 2 following reasons.On the one hand,climate change presents an urgent global challenge with a recognized need for collective action,where AI affords a tool thatbuilding on existing efforts,data sources,and machine learning technologycould augment earth observation,energy management,and other f

14、ields important to meeting the climate challenge.On the other hand,privacy-enhancing technologies(PETs)are a budding technology field where collective resources can accelerate development to help overcome barriers to data access due to concerns about privacy,security,and ethics,as well as proprietar

15、y interests and economic protectionism.Together,these subjects present a balance between applying existing AI to tackle pressing global issues and expanding the frontiers of AI in ways that promote responsible use of this powerful combination of techniques.Criteria for selection of projects To ident

16、ify potential subjects for global R&D initiatives,our October 2021 report proposed five criteria:Global significance,global scale,the public good nature of the projects subject and goals,its deeply collaborative nature and the need for its impacts to be measurable and assessable.Feedback from FCAI p

17、articipants substantially validated these criteria but demonstrated that they have diverse facets,synergies,and overlaps that add complexity to choices.Furthermore,discussion among participants clarified intersections and contrasts among these suggested criteria that provided a sharper lens to exami

18、ne differing choices of areas for international R&D collaboration.1.Global significance implies that such a project should be aimed at importantglobal issues that demand transnational solutions.The shared importance of theissues should give all participants a common stake and,if successful,couldcont

19、ribute toward global welfare.As emerged from discussion with FCAIparticipants,global significance has several different facets,encompassingimpacts on humanity or the environment(“people,planet,and prosperity”).Another closely related dimension is urgency,which refers to the opportunitycosts of faili

20、ng to act.From this latter standpoint,climate change and globalhealth stand out as obvious priorities.On narrower planes,however,it isimportant to consider dimensions such as the impact on AI innovation,theprojects impact on international cooperation,and its impact on specific fields ofresearch.Ulti

21、mately,the selection depends on balancing priorities.2.The global scale criterion expresses that the scope and ambition of the projectrequire resourcesfunding,access to data,computing power,knowledge,talent,and financial resources to support theseon a large enough scale that thepooled support of lea

22、ding governments and institutions adds significant value.Such resource needs correlate in significant part with the magnitude of theglobal problem targeted.Nonetheless,this magnitude does not correlate directlywith project complexity:Climate,health,and economic measurement all involvemassive complex

23、 systems,but data and computational methods to addressclimate are more available in the present than they are for global health oreconometrics.By contrast,PETs involve a much more discrete set of problemsbut highly complex computation.Again,scope and ambition must be balancedwith feasibility.We reco

24、mmend that FCAI governments,stakeholders,and other likeminded entities prioritize cooperative R&D efforts on deployment of AI as a tool for climate change monitoring and management and accelerating development and adoption of privacy enhancing technologies.GLOBAL AI COOPERATION ON THE GROUND|AI RESE

25、ARCH AND DEVELOPMENT ON A GLOBAL SCALE Brookings Institution 3 3.The public good nature of the project means that,if successful,the output of theproject would benefit the global community at large.Both the project and itsresults should be available to all participants as well as used to improve acce

26、ssto data,talent,and computing capacity in less developed countries.A publicgood should provide wide benefit and be widely available,inclusive,and non-rivalrous.Hence,this criterion requires assessment of who benefits and,withsignificant R&D and data sources in private hands assessments of the priva

27、temarketplace,how to incentivize private sector participation.All the subject areasdiscussed have significant public good aspects;some feature significantamounts of needed data in private hands(especially from the health andpayments sectors,though varying in degree by country).4.The collaborative na

28、ture of the project means that the project is likely tonecessitate a multi-disciplinary approach addressing regulatory,ethical,risk,andtechnical questions in a context that is concrete and in which the participantshave incentives to achieve results.It would amount to a large-scale,shared,andinternat

29、ional sandbox or test bed for AI methods,governance,and regulatorymodels.Our series FCAI dialogues in 2020-2022 led us to observe that abstractissues of regulation and ethics are addressed most effectively in the context ofconcrete use cases,and a shared commitment to seek concrete solutions wouldpr

30、ovide incentives to resolve differences.Hence,appraising this criterion meansconsidering how strong incentives for collaboration are:How much do the globalimpact or public good nature of the project motivate participants to collaborate;and where may this motivation be reduced by inevitable differenc

31、es in policies,culture,or resources?5.Finally,the impact must be assessable.Any project will need to be monitored andaudited commensurately with its scale,public visibility,and experimental nature.Participants will need to assess progress toward both defined project goals andimpact in addressing eth

32、ics,risk,regulatory issues,and other,broader goals.Lessons from the projects should be shared widely.As pointed out by FCAIparticipants,this is an especially cross-cutting criterion.Here,feasibility entersinto consideration,as well as an assessment of how AI can contribute value.Ourprogress report a

33、lluded to the spinoff benefits of scientific programs like theApollo program or the Human Genome Project;6 these are collateral benefits ofprojects with very defined objectivesgoing to the moon and mapping the entirehuman genome,respectively.The targeting of areas should not engage in“techno-solutio

34、nism,”the assumption that because a subject is big and complex,AI necessarily can and will find solutions.To the contrary,in most fields this willbe at most a tool in the context of multi-pronged strategies.The potential impactof AI on some subject areas is more defined,on others more speculative.Su

35、bjects with more established datasets and pathways for data access anddeployment of machine learning offer greater probability of definable impact.The choice of most highly deserving areas for collaborative R&D projectstherefore involves important balancing of factors,for example between seekingto m

36、ature emerging technologies versus exploiting readily available technologiesfor immediate public benefit.These trade-offs result from technical issues.Different subject areas will present different issues such as what data andcomputing capacity is needed,what is available,and applicable data governa

37、nceGLOBAL AI COOPERATION ON THE GROUND|AI RESEARCH AND DEVELOPMENT ON A GLOBAL SCALE Brookings Institution 4 considerations.For example,NASA,7 the European Space Agency(ESA),8 and meteorological agencies9 make public vast databases of observations,while sharing of health data is subject to instituti

38、onal and privacy constraints.Similarly,different subjects call for different AI methods,and some require greater or more disruptive innovation to achieve results.As discussed in our progress report,differences in data formatting and metadata may present obstacles to sharing data sets even where thes

39、e are available.What consideration should be given to availability of data and proven AI technology?What priority should be given to expanding the boundaries of AI versus impact on targeted global problems?6.Governance is a potential additional criterion not discussed in our October 2021report.Effec

40、tive governance will be necessary to the success of large-scalecollaborative projects.The extent to which the product of joint R&D is a publicgood will depend on governance rules,among other factors;so will the operationof any kind of international regulatory sandbox and the call on resources tosupp

41、ort a project.The progress report pointed to CERN,the International SpaceStation(ISS),and HGP as potential models.The Appendix below provides ageneral summary of the different governance structures,funding models,andoutcomes of these entities,which range from a body of governmentrepresentatives over

42、seeing a structure of research contributors(CERN),to ahandful of government agencies(ISS),to an aggregation of research institutionssupported by government grants(HGP).These models provide starting points forconsidering governance.This range of variation along with common sensenevertheless suggest t

43、hat governance is likely to depend on the nature of theR&D and the stakeholders and funding sources involved.It follows that anyextensive discussion of governance should follow from discussion of subjectareas for R&D.In addition,governance could be affected by other subjectsidentified for deeper FCA

44、I discussion,especially topics on regulatory alignment,including coordination on regulatory sandboxes(recommendation R7).For thesereasons,we did not address governance in our February 2022 roundtable or inthis paper.A group of scholars from the Stanford Institute for Human-CenteredArtificial Intelli

45、gence(HAI)have a proposal that resonates with ours in proposingan international AI research institute but take a different tack in beginningstructure by recommending that the United States take the initiative to establishsuch an institute that would resemble CERN.10With the background of this refine

46、d understanding of selection criteria,the February 2022 roundtable turned to specific areas of potential international R&D collaboration discussed below and the opportunities and challenges involved.GLOBAL AI COOPERATION ON THE GROUND|AI RESEARCH AND DEVELOPMENT ON A GLOBAL SCALE Brookings Instituti

47、on 5 Assessing areas for R&D collaboration In this section,we first provide a brief look at the four proposed areas for large-scale collaborative R&D projects:Climate change,public health,privacy protecting technologies,and economic measurement,with a focus on AI-related developments and policies in

48、 the field and existing efforts to develop international cooperation.We then turn to a comparison of these subjects through the lens of our analysis and discussions with FCAI participants.Finally,we explain why this comparison led to climate change monitoring and PETs as prime prospects.Overview of

49、potential projects 1.Privacy-Enhancing Technologies(PETs)PETs may enable some of the benefits of data sharing,cross-border data flows,and open government data while protecting data privacy and compliance with data protection laws and other restrictions on data.This category of technologies is broad:

50、It can encompass technologies for secure communications and transactions such as secure messaging and other applications for encryption;tools for privacy management such as anonymization and pseudonymization of individual records and other data governance and compliance;software to prevent or limit