加勒比地区的伦理、平等与人工智能

2021 年 6 月 22 日

人工智能(AI)将影响全球各国的社会经济发展。加勒比海国家尤其容易受到影响,因为它们往往是劳动密集型经济体,因此面临着自动化和人工智能对经济和社会造成重大破坏的风险。该领域的三位专家--科德尔-格林(Cordel Green)、斯泰西-拉塞尔(Stacey Russell)和埃里卡-西蒙斯(Erica Simmons)--讨论了这些问题以及更多其他问题。

加勒比地区的伦理平等与人工智能 Spotify 链接 加勒比地区苹果公司的伦理平等与人工智能链接

PIA-MILAN GREEN: Welcome, everyone, to our virtual panel, "Ethics, Equality, & AI in the Caribbean." My name is Pia-Milan Green, and I am a research fellow for the Artificial Intelligence & Equality Initiative (AIEI) of Carnegie Council for Ethics in International Affairs. This Initiative seeks to understand the innumerable ways in which artificial intelligence (AI) impacts equality and in response to propose potential mechanisms to ensure the benefits of AI for all people.

Today we have brought together a fantastic group, and I will start quickly by introducing the panelists before we jump into some questions.

Ladies first. Our first panelist is Stacey Natasha Russell. Stacey, a native of the Caribbean island of Barbados, has had a 26-year career in journalism and media communications in the Caribbean and in the United Kingdom with a passion for promoting awareness of artificial intelligence, cybersecurity, privacy, and data protection. She holds a Bachelor of social science degree in communication, a Bachelor of law degree and a Master of law degree in media law, policy, and practice. She currently counsels a cross-section of digital end users in the Caribbean in online safety and security through her association with the U.K.-based charity Get Safe Online.

Our next panelist is Erica Simmons, a global technologist, innovator, catalyst, and philanthropist with Fortune 100 business experience. Erica's professional career has included a successful 25 years in the enterprise software technology industry, holding strategic roles for Oracle Corporation, Expedia, and one of the world's largest engineering companies, Siemens AG. Today she is the executive director for Jamaica's first Fourth Industrial Revolution technology research & development center, located at the Caribbean Maritime University in Kingston, Jamaica. She is also involved with the Institute of Electrical and Electronic Engineers (IEEE), one of the world's largest professional technical associations and standard-setting bodies, where she serves as the 2021 chairwoman of the IEEE Jamaica.

Lastly we have Cordel Green. Cordel Green is an attorney-at-law, a former assistant attorney general of Jamaica, and a former broadcaster. He is currently the executive director of the Broadcasting Commission, which regulates radio, television, and cable in Jamaica. Cordel is also the vice-chairman of the international bureau for the United Nations Educational, Scientific, and Cultural Organization's (UNESCO) Information For All Programme and chairman of the Working Group on Information Accessibility. His other affiliations include being a member of the advisory council for the Carnegie AI & Equality Initiative, and he is the commissioner of the Jamaica National Commission for UNESCO. Cordel holds a Bachelor of arts degree, a Bachelor of law degree, a Master of law degree with distinction, a Master's degree in business administration, and is an alumnus of the Harvard Kennedy School Executive Education Program as well as a Chevening Scholar.

Welcome.

Cordel, I think we will start with you. Some persons consider artificial intelligence to be beyond the reach of developing nations, in particular Caribbean islands. How would you respond to such an assumption?

CORDEL GREEN: First of all, Pia-Milan, thank you very much for your kind introduction, and thanks to the Carnegie AI & Equality Initiative for having us here. May I put on the record that there is clearly a gender imbalance, but I have absolutely no complaints. I love it, and I love being a minority.

Now to your question. I think it is not only a dangerous assumption to say the Caribbean and developing countries are not ready to grasp AI. It is flatly wrong. First of all, speaking for the Caribbean, Caribbean people are early technology adopters, so technology is widely diffused in the region. AI will be no different.

The question is whether we will only be the consumers or the innovators, and we are naturally innovative people, and I know for a fact that we have very impressive technology innovators who are doing great work. I just discovered one, Eldon Marks in Guyana, who is coming very soon with a game-changer to the chatbot experience. The weakness is that there is no Caribbean Silicon Valley to take our innovations to the next level.

We need to rethink our development priorities and strategies, both from the point of view of what we set aside in our national budgets for research and development, and as importantly, what our developmental partners choose to prioritize when they give us developmental aid. I find it very odd that with all the revolution that is taking place, when you look at the assistance being given to developing countries such as Jamaica, capacity to harness exponential technologies such as artificial intelligence does not loom very large. I find something inherently inequitable and flawed in the notion of development that is crafted for our region by ourselves and by our partners.

My point is that the question is not whether AI is beyond the grasp of the Caribbean. It simply cannot be. The Caribbean cannot afford to miss this wave of development. But I call for an international framework which is equitable, and there is no question in my mind that the existing global trade, human rights, and multilateral framework within which this general purpose technology is operating is not stacked in favor of the Caribbean. We have a lot of deconstruction to do there.

I will pause there and see whether my colleagues have anything to say on this, and then I might have something else to add.

ERICA SIMMONS: I will jump in and totally support what you have said, Cordel. The research that we have completed over the last few months has validated that. We know that there is exponential technology on the market. We know we have developmental challenges here in the Caribbean that we are trying to overcome.

You are right, it has been interesting. We have not found a lot of the technologies and the solutions focused on our real developmental challenges, so I would agree with you. Part of that is maybe we need to become more active in the community, because that was another thing that we noticed. There is not a lot of activity even on the global stage from the Caribbean when we are dealing with artificial intelligence.

For example, in the standards bodies we talked about the Institute of Electrical and Electronics Engineers, which is a large standard-setting body. There are some initiatives that are setting standards for ethics in product design, and when I looked at that group there was no representation from the Caribbean. So, of course, I have inserted myself into the group now so that we can at least have a talk, at least there is a conduit to take the information that is happening at the global stage and bring it here to the Caribbean.

Then we have to get a lot more interested in what's going on. We have to get a lot more educated on the possibilities of these technologies. That's what I would add. I am agreeing with you, and I think it is a two-way street. We have to demand that we get a seat at the table, and we have to bring our seat sometimes and pull it up, and then we also have to be interested here in the Caribbean on solving problems, on embracing the technology and the digital transformation that is happening around the world. How do we bring that to the region is what we have to focus on.

CORDEL GREEN: One positive thing out there is that there is now a very important doable conversation taking place within UNESCO on the UNESCO's recommendations on the ethics of artificial intelligence. This is happening at the same time that the European Union is promulgating its AI Act and the IEEE and others are developing standards. I am very pleased to say that in that global debate there is a very strong Caribbean voice, particularly Jamaica, Cuba, Grenada, and St. Lucia being rule makers and not just rule takers.

I should mention as well something that both Erica and Stacey are very familiar with because they were key contributors, the Broadcasting Commission of Jamaica, in partnership with UNESCO, launched a Caribbean Artificial Intelligence Initiative, which has resulted in the crafting of a Caribbean artificial intelligence policy brief.

So we are taking AI seriously. We are not just sitting by as observers of global policy. We are working to influence it but clearly, as Erica said, there is more to be done, but we are definitely not a set of people who see AI as beyond our grasp, even if others are thinking that way based on past experience when we really have not influenced as significantly as we should global developments, which is very odd because the Caribbean has a global, cultural, and intellectual footprint that bears no relation to our size and economic ranking.

Four Nobel laureates. When the Rhodes Scholarship had to celebrate its centenary, one of the four persons they could find globally to represent the intellect and ethics of the Rhodes Scholarship was the late professor Rex Nettleford, and we have gifted the world with Rastafari, reggae music, and one of the greatest acoustic instruments of the 20th century, the steelpan from Trinidad and Tobago. So there is no question at all that in the realm of innovation we should be upfront. That is an anomaly that has to be corrected, and the primary responsibility rests with Caribbean people. That is why we have developed the Caribbean AI Initiative, and hopefully it will be promulgated, and we will take our rightful place and ride the AI rising tide that is to lift all ships.

STACEY RUSSELL: I am definitely agreeing with you, Cordel, and Erica. You know, Cordel, what you just said there about us pulling ourselves up, and Erica, what you said earlier about us not waiting to be called to the table but taking ourselves to the table, that is what I would really like to see a lot more of in the Caribbean in relation to artificial intelligence and the use of technology generally. I think it has to start in the schools. We have to have grassroots programs to reach our young people, for the young people within homes and families and communities to then educate the older ones who tiptoe around the progression in our use of technology globally. Those are the things I am passionate about.

How do we develop industries here so that we do not typically go to school to become a lawyer or a doctor or a nurse or other trades, but what are we actually doing to incorporate in the fabric of our education and our industries artificial intelligence?

PIA-MILAN GREEN: I think those are salient points, and I believe they underscore the very purpose of our panel, which is to bring increased visibility to the field of AI in the Caribbean, so thank you all for that.

Erica, as I understand it, you just worked on the UNESCO Caribbean AI Policy Roadmap. Can you share with us a little bit about what you learned from that experience?

ERICA SIMMONS: Yes. It was very exciting to be a part of that experience, Pia-Milan, and to really develop my understanding and help the region and all of us to develop our understanding of where we are in the world in terms of our AI adoption, our use, and our education.

I will just mention a few learnings along the way. I have mentioned one before, that we definitely are experiencing AI on a daily basis here in the Caribbean just because the technologies are being deployed on our cellphones and we are interacting with Netflix and Spotify and all of those products that we interact with on a daily basis. Citizens are interacting with AI. What we are not doing though, is the development work, working on solutions for AI. Not a lot of companies—I think Eldon Marks is one of those people who has a company focused on AI, but when you look at the region, we are not very focused on it as a product offering.

The second thing that we noticed is the education, as Stacey just talked about, the awareness. I want to congratulate Cordel and everything that they have done in the Broadcasting Commission of Jamaica. They are probably one of the countries that are out front in the evangelism of AI, educating the citizens on the possibilities available, so good work for the Broadcasting Commission of Jamaica, but we definitely need to see the other countries take up media and information literacy regarding artificial intelligence. We are not seeing a lot. Maybe some of you all can adopt the best practices that are going on in Jamaica with the Broadcasting Commission.

The third big point, Pia-Milan, is just the lack of information on the topic from a research perspective, so looking for researchers. We have wonderful universities in the region, as Cordel said, scholars across all kinds of topics, but not a lot of research in the area of artificial intelligence. Even in the areas where we could use it for benefit, like our climate change issues that we have—how do we apply the technology to assist us with prediction, or even after a hurricane comes, how do we marshal forces to be able to respond in a resilient manner when the data is not there?

So we have to start increasing more in the university level, implementing, and talking about it and writing about it. There were fewer than ten academic papers written in this area in the region, so most of the research that we were doing with the paper really was outside of the region, seeing what the best practices were, what's happening in other parts of the world, and trying to see how we were going to do it here in Jamaica.

Another profound point that I would like to make is the fact that we don't really have a coordinated approach to artificial intelligence and even general technologies in the region. We are taking it island by island, and we are trying to go it alone and individually. Obviously we are too small. It is not a scaleable approach, so we need to see how we are going to collaborate a lot more across our islands to be able to really make something out of artificial intelligence and the opportunities ahead.

We have to invest, though, Pia, because to be honest we don't have the infrastructure that is needed to support artificial intelligence in the region. We need processing power in the form of graphics processing units. We need cloud offerings located in the region and not in other places. Our countries in the region should look at coming together to see how we are going to build the infrastructure that will form the foundation for building other solutions on top of it, but we cannot be in the game unless we have those technologies resident in the region.

Those are just some of the findings. I think we have a very exciting opportunity for us in some of our industry sectors, such as tourism. Tourism and artificial intelligence would be a great specialization for us in the region. We have the domain expertise here. Everybody knows it's a great place to come and visit. How do we take that domain expertise about tourism, join it with this general purpose technology, and become a leader in those types of solutions for the market? So if people want to come up with AI-based technology products or they want those solutions, they should be coming to our region. We know the product. We know tourism. So how do we support those industries that we are already in with this technology is something we talked about in the paper, and then, how do we look at new sectors that we can tackle, like e-sports? That was a big one.

We are great in the physical realm in terms of sporting. Everybody knows we have the fastest man in the world and the fastest woman in the world, so we know something about that. How do we take something like a new sort of spin, e-sporting, and take our domain knowledge and then transform it into a new industry sector? Remember, we are going after wealth generation here in our region, so we have to after the new. It is better possibilities for us to generate wealth when we are looking at new sectors.

I was thrilled to be part of the research team and bolstering up my knowledge in this area of artificial intelligence in the Caribbean. I am extremely encouraged, though, Pia. For all the challenges that we have and have had, I think the research shows that we have great possibilities and we have wonderful talent in the region. And, as Cordel said, if we can get some more investment and bolster up our education, like Stacey said, we will be in a great position. So we need to look at the positive part of this and try to go forth.

CORDEL GREEN: Erica, just hearing you talk gets me excited about the possibilities. For me, one of the shocking things that came out of that whole process of engagement was a snap poll of stakeholders which revealed—it certainly shocked me—that only 12 percent of businesses in the Caribbean were using AI and that we had low adoption even among large digital leaders in the region. It wasn't a scientific poll, but coming out of a meeting of private sector, public sector, and AI-engaged individuals I didn't realize that we had such a weak base on which to build a regional offensive.

I am just underscoring in this discussion the need for urgent action at the policy level, at the business level, and at the investment level. At 12 percent engagement, and largely that would be in the predictable areas—the financial sector, chatbots for customer service, and so on—that is not very impressive. I am hoping that it is because we are not doing adequate research, which explains that low number, and that when we start to dig deeper we will have a better profile in the AI use in business in particular.

ERICA SIMMONS: I just want to follow up. Remember we also heard from Dr. Henkel Valentine, who told us about the research he was doing with prostate cancer cell lines in the Caribbean and that he was using artificial intelligence to do that. So it is a beacon that we have the capability, we have the medical expertise, and I also think the healthcare sector is a wonderful sector for us to go into.

I just want to say: COVID-19. COVID-19 has had a significant impact on us in the region, but really what it has done as well is it has advanced the digital transformation. So even over the last year we have seen significant improvements in the banking sector and our ability to interact there. A lot of people are going online now. We are able to pay for things using digital applications.

So we are the beginning I do believe of a transformation of our region if only we can believe and if only we can push towards that direction. So the businesses have to definitely be more in tune with using this technology because it's for efficiency. We are talking about cost savings, efficiency, and more revenue. We need our governments to make sure that they have policies in place to support that because businesses are going to be motivated by incentives, and then we need our education sector to get involved at all levels, from primary school level so the kids can understand that when they are online they are interacting with AI and AI is learning about them, so they can monitor and have their behaviors be appropriate for the environment, all the way up to our research and development universities implementing it to find problems. Those were some things also that came out in the study. We are doing great things. We just need to accelerate it now.

PIA-MILAN GREEN: Thank you so much, Erica. Fantastic points. I am actually going to come back in a little while to the point that you made about a collaborative effort in the Caribbean.

What we are hearing is that next-purpose technology has already started to and will disrupt all of the sectors in the Caribbean—our approach to manufacturing, content creation, consumption of content, agriculture, and the creative industries.

Stacey, as we discuss these policy roadmaps that we have started to create, how do you feel that the Caribbean people will view the ethics of using artificial intelligence generally? What would the ordinary man's response be to that?

STACEY RUSSELL: It's interesting that you are asking about a general context. I always find it so much easier when we can relate to a particular example, a particular sector, a particular way of doing things, and then we can extrapolate it to what it might mean for us in general terms.

When I was considering this forum and the equality aspects of it, I started to think about an area that will touch on the lives of each and every one of us, and that is crime, that is policing, that is keeping our societies safe. It is about what our governments will do. Here we are talking about industry, but the government has a big role to play in all of this in terms of regulation and in terms of leading the way in certain departments.

I think policing is one area that I can definitely focus on in terms of giving an example of how we can be affected in other areas of our lives. In looking at this example it is about the ethics, the rightness, the moral principles as we understand them in the Caribbean, of incorporating these various computer systems that are programmed to perform tasks into the fabric of law enforcement.

It is generally agreed that the use of AI in policing is intended to make forces more efficient and more effective as they strive to keep our societies safer, for example, automated monitoring, like the use of closed-circuit television that a lot of us might not like having our public persona captured when we are moving around in our daily lives, but it is there for a purpose, in case anything goes wrong. We might be the law-abiding citizens, but then there might be others out there who are not doing the right thing.

Then, of course, when officers wearing their body cameras can capture what actually happens when they are interacting with members of the public, that is an example of using artificial intelligence in policing for the benefit of the masses. Once it is done and executed in a way that is responsible I would say that, yes, it is the right thing to do, it is the moral thing to do, it is the ethical thing to do to ensure that we are all kept safe and to ensure that officers are not abusing their powers.

AI can be used for the better in terms of reducing the use of paperwork for officers, to make their work more productive, processing big chunks of data like those from suspects' devices in digital crime, and somewhat controversially, being able to predict who will commit crimes in the future by using arrest histories and other AI-driven data. That is something that I can talk a bit more about.

For instance, something that comes to mind is an example where in late 2019 police officers in Pasco County, Florida, turned up at a home looking for a teenage boy about 8:00 at night, and they insisted that they needed to check his room, saying that he was a high-category criminal and was being watched by police. His family knew that he was on probation for one offense that he committed in his life. He was about 14 or 15, and he and his friends had stolen some motorized bicycles, but for that reason, having information about him on file, using that as an example, an AI algorithm actually said to the officers that this minor may commit future crimes based on what he had done in the past.

These are some of the issues that we now have to consider for the Caribbean and how ethical it is because many of us might want to support "predictive policing," as it is called, with this notion that police can be so proactive in using the technology to crime-proof our societies, but we also have to consider that incorporating this type of AI into Caribbean policing could easily paralyze underprivileged communities in particular. It can lead to greater social unrest than the predicted crime could induce. It could wipe out a young generation that would have no way back from one childhood mistake. This example from Florida in the United States indicated that the particular county had a mandate to utilize the technology in this way, and over five years they targeted about 1,000 citizens, and one in ten of them were below the age of 18.

These are some of the ethical issues that we have to consider as we want so much for artificial intelligence to drive what we do, who we are, how we feel, how we operate, and how we interact as a people.

There is more that I can talk about.

CORDEL GREEN: Stacey, I wanted to join you on that question of the trust before we move on because that is the elephant that I think needs to be put smack in the middle of all of this, and the technological possibilities.

We do have a problem in the Caribbean. I don't think we are a minority. It seems to be more global. We have a huge trust deficit now in our institutions, particularly government.

You are quite right to signal law enforcement. Too much of our national budgets go to policing and law enforcement, and our judicial system is clearly inefficient. But we are not going to trust these algorithms that are being created and architectured in other places with all kinds of in-built biases, having gone through the horrible experience of slavery and colonization and neocolonialism.

I think we are right back to where Erica took us. We cannot be isolated, but we have to create our own solutions, and in these very sensitive areas we have to resist technological determinism. Things don't have to be the way they are. They don't have to be designed the way they are designed, and we do not have to take external solutions—which, by the way, I am not saying you were saying. I am just joining you in agreeing that we need to address these critical areas and use the technology.

But the approach we take is very important, and I think in areas of solving judicial problems and law enforcement, I am far more comfortable with a Caribbean-built and purposed solution than anything that is coming off the shelf because that American experience that you just alluded to I believe is more commonplace in the experience of black people in the United States of America and can be generalized. So I am joining you in what the technology portends, but I am seeing a lot of ethical and policy challenges that we will need to address in a uniquely Caribbean way.

ERICA SIMMONS: I would just add to that that we have to develop a strategy for the sectors that we do want to implement the technology. The policing, although we have a huge problem, is going to be an ethically challenging environment for us to work in, and I think that there are probably other low-hanging fruit areas that do not allow the algorithms to make those sorts of life-and-death decisions about us that we can probably implement better in terms of, like you said, making our government more efficient.

I totally agree with you, Stacey. The government sector in the Caribbean has to drive the digital transformation. They really, really do. From that you will get private-sector lobbing on, but I do think we have to decide where are we going to work with what technologies because we can pick a hotbed area such as policing and never get off the mark because that area is such a contentious area for a lot of reasons. I actually think that there are other low-hanging fruit areas that we can focus on and maybe look a little bit long term on how we are going to implement AI in those areas of law enforcement and policing.

I would like to add, though, on the judicial system, I read today in the newspaper here in Jamaica that we have a backlog of people waiting to even get seen by a judge to get bail on crimes that they are accused of. Maybe in that area we could use the technology to work through whatever paperwork we need so that we can make our judicial system a little bit stronger. If we wanted to focus, maybe that is an area that would not be as contentious as facial recognition and the policing side of it, although we need to solve our crime problem. I am not advocating that we don't.

STACEY RUSSELL: Yes, Erica and Cordel. I get what you are saying 100 percent. I think we need to be up for whatever challenges are here.

Erica, you are saying we have a big crime problem in the Caribbean in so many of our islands. Just yesterday a wanted man in Barbados was gunned down because he pulled out a gun. Police had to defend themselves, and he is now dead.

The issues that we do need to consider are transparency and trust. It is about being able to balance the benefits and the risks of what we are trying to achieve.

Also we have to think about data protection and the masses of information that we are collecting, ensuring that we are not truly infringing on people's right to privacy and so forth. But I would not say that we should not be thinking ahead. I understand about having to start somewhere, but we also have to be thinking ahead so that we can eventually forge ahead in terms of being on top of our social challenges in the region.

PIA-MILAN GREEN: Thank you all. I think something I am hearing a lot is this idea of a Caribbean approach.

Stacey, I am going to probe on that for a minute. Barbados Prime Minister Mia Mottley has indicated that in the context of AI she believes that no one should be left behind. But this raises the question of how. How do we achieve this in the Caribbean?

One perspective that I actually came across is from a former dean of business at the University of Technology in Jamaica, Professor Paul Golding, and he speaks of a "regional integration ecosystem approach" to AI development and governance as opposed to countries responding independently.

Stacey, let me ask you: Do you believe that that perspective is the way to achieve Prime Minister Mottley's vision of no one being left behind?

STACEY RUSSELL: My very short answer is yes because we always have strength in numbers, wherever you want to start, if you want to start from the family, you want to start from the community, or you want to start with businesses working together. We have to work together as a region to achieve these things and definitely by making our people the center of what we are going to achieve. We have the Caribbean Community (CARICOM) single market and economy, which I think we don't particularly call upon with the vigor and vitality with which it was articulated in the Treaty of Chaguaramas in terms of achieving real practical goals for our people.

But if we are to even look at this pandemic and look at other natural disasters, for instance, the volcanic eruption in St. Vincent and the Grenadines, what is happening with flooding in Guyana, and how we are understanding that we do need to band together and to help each other in the region for the good of the people. That is definitely the way to go in relation to implementing strategies, regulation, education, investment, and so forth where artificial intelligence is concerned.

CORDEL GREEN: I want to join in and support that point of view and to emphasize in particular data. I really think if we don't emerge as a data bloc we are going to be very much insignificant in this domain.

That is one of the things I particularly liked about the process that led to the Caribbean AI Roadmap because this has been emphasized and reflected in the Roadmap, that the region's most valuable renewable resource is its human creativity in the form of thought data and cultural data. I personally think that is precisely what is behind the European data protection rules, the General Data Protection Regulation (GDPR). It is framed as a human rights construct, and I support that entirely.

I actually think it is a trans-Atlantic trade tool. The United States of America and China are ahead in terms of research on artificial intelligence technology and the deployment of AI solutions, but the oil for AI is data. I think that is what the Europeans recognize. That is what GDPR allows, for the Europeans to protect their oil, and I believe that we are going to have to do the same thing as Caribbean people. Our fortunes are not the same. The extraterritorial reach and impact of their GDPR—we can never even dream of doing that. But even for our own internal use and exploitation, we need to corral Caribbean thought and cultural data, and I think that is probably what Prime Minister Mottley was alluding to, and I entirely support that notion.

PIA-MILAN GREEN: Yes, and I think, Erica, if you could talk to what that approach might look like. How can we garner the support and the activism of ordinary Caribbean people in policy-setting rules, and then how would that ecosystem approach translate? What would the islands need to do?

ERICA SIMMONS: I just want to say I completely agree with everything that Cordel said. The data is very, very important. To that conversation we have tons of existing data. We have a wealth of existing data right now that is not in a format that can be used by these AI systems, so we definitely need to get better at the data part of it. We have to understand that we are creating data every day, and we have to become expert at transforming what I would call analog-based data, data that is unstructured, into a way that we can package it, sell it, and license it perhaps.

We have lots of wealth of data on every island. Every island has someone who is a curator of their national histories, assets, museums, and other things. So we have data. We have to become expert at it.

But as far as a strategy, here is a thought that we could potentially go with. I say that "we cannot be an inch deep and a mile wide." That is what I like to say, meaning every island cannot know everything and cannot have an expertise in all of artificial intelligence.

We have 30-something islands in the Caribbean, or however many we want to focus on. We need to divide our islands up and give every island a specialization. For example, there could be one island that is all focused on the ethics of artificial intelligence and really understanding that and being able to give us guidance and solutions and systems from an ethical perspective. We might have another island that is leading the way on our data strategy. We need a data strategy. How are we going to collect this data that we already have? How are we going to package it? How are we going to sell it? How are we going to leverage it in these new systems that we are creating?

The other thing that we need to do definitely is to make sure that we are looking at our infrastructure again. Every island cannot get everything that they need to support the AI infrastructure. For example, could we go in with a group perspective to procure some resources, some infrastructure that would be deployed in our region? We could spread this out because we know we have hurricane challenges and natural disaster challenges, so we cannot have all those resources on one island. We need to have a network effect where we can spread the resources throughout the islands. Then we can have each island or a cluster of islands—it does not have to be one—have a specialization in something related to artificial intelligence where the whole island, from the elementary to middle to high schools and universities are focused on becoming very, very expert at that.

That I think hopefully is a way to move forward because this area of artificial intelligence—and remember, artificial intelligence is just one of the technologies that are transforming our future, so we cannot know everything about it. So we have to dissect it, assign islands specializations, and have them focus on that. I truly believe that is probably a way forward, and that speaks to the whole collaborative aspect of it. Each island should not be doing this individually.

So I am thinking that might be a way that we can work together. Let's have each of our islands have specialties, and then we can teach each other.

What do you think about that, Cordel and Stacey?

CORDEL GREEN: Yes, yes, yes, I am with you 100 percent, and we are already specialists in one way or another, so that can be quite organic.

STACEY RUSSELL: Yes, and bringing it under the body of the CARICOM Secretariat would be the way to go. We already have various ministers responsible for various things, heads of government in each one of the CARICOM member countries responsible for various areas. There is also a rotation of who is the chairman of CARICOM every six months of the year. Definitely I would say that is the way to start. That is the place to start. We have a framework. We just have to action these things, and we have to be willing to work hard and set goals and timelines to make them happen.

CORDEL GREEN: And mobilize the people. There is nothing like a crisis to mobilize people. I think if the Caribbean people figured out how much of their data is being captured, colonized, commercialized, and making Big Tech companies very rich, you might be surprised at how much Caribbean people would rally behind some of these ideas, Erica's low-hanging fruits, at the very least creating data and investing in data banks.

And we are passionate about these things. I just wanted to add that, that yes, the CARICOM structure, but the weakness with the CARICOM structure is I don't believe there is sufficient engagement on the ground, and with a technology such as AI, given its nature and how transversal the effects are, we are a labor-intensive economy, and so there are implications for automation and the introduction of artificial intelligence. We definitely need to get the people involved in the discussion and in the solution, and I think they are ready for it and can help to shape the process.

STACEY RUSSELL: I did want to go back to the whole notion of policing as an example of how we employ artificial intelligence because we have to also think about our women and what part do our female folk in the Caribbean play. What is really meant by "equity" and "diversity" when we talk about these issues? How does artificial intelligence affect women, for instance?

We are reading headlines every so often, whether they be here in the region or whether they be from other parts of the world, that women need help in so many ways. The biggest thing that has been talked about recently is women being harassed and women being victims of rape. We had this incident in London, where a serving London Metropolitan Police officer, someone who is supposed to protect, serve, and reassure, just last week confessed to abducting a woman and raping and killing her. You have just today ministers of government in the United Kingdom apologizing and saying that they have failed women in this regard of policing and protecting them whilst it has been on their watch.

Recently in Barbados you had a woman speaking through her attorney about being manhandled by a police officer who came to complete a search at her home and being verbally abused by him. I know that it might not be one of the short-term issues that we want to go after, but it is fine for us to think futuristically. It is okay for us to think, Well, maybe we are not going to achieve this next year or in the next two or even the next five years, but let's start having a conversation about things to make it happen.

For instance, we can start thinking about how we can use a chat box when police officers are investigating crimes. Erica was just talking about reading about cases languishing in the courts, but that is where the work starts. If it doesn't come through the chain, come through the channel of investigation, of taking witness statements, and information sharing between the two sides, and so forth, how far can you get in terms of the case coming before the magistrate or the judge?

CORDEL GREEN: I want to agree with you, Stacey. I know we have four minutes to go, but I need to just say one thing, though.

I really caution—I accept what AI portends and so on. I just don't believe some of these things require technology solutions or that technology will solve these things. I entirely agree with you that these are serious social issues we are having with how we treat our women.

This is one of the moments when I get back to human centricity. We need to go back to some core principles about who we are as a people because unfortunately I think if we don't address those things, some of the very biases and dysfunctionalities that we see playing out in these relations will find their way in whatever technology solutions we devise.

In short I see it as a combination, and there is a critical human element aspect to this which needs attention, and that is my takeaway from what I have heard you say. In addition to looking at the technology solutions, we also have to balance it with remembering that we are talking about human beings and put human beings at the center of whatever developmental architecture we devise, our own artificial intelligence.

ERICA SIMMONS: Pia, could I just add that we are in a perfect position for this because we are the melting pot of the world here in the Caribbean. We have people who speak multiple languages. We have different climates. We have different backgrounds. We have women and children. We really have a great position in our cultural diversity here in the region to be able to think big and strategically about AI and the challenges that we have with AI.`

PIA-MILAN GREEN: What a point to end on. Thank you all. Thanks again to our fantastic panelists and to everyone who tuned in today.

Before we sign off, a reminder that the full video and transcript of today's panel will be available on Carnegie Council's website at carnegiecouncil.org.

Be safe, everyone. Thank you.

预测性人工智能技术的伦理意义是什么?

您可能还喜欢

2024 年 7 月 31 日 - 播客

负责任的人工智能与大型模型的伦理权衡,与 Sara Hooker 合著

在本期节目中,高级研究员安雅-卡斯珀森(Anja Kaspersen)与 Cohere for AI 的萨拉-胡克(Sara Hooker)讨论了模型设计、模型偏差和数据表示等问题。

2024 年 7 月 24 日 - 播客

人工智能与战争:军备控制与威慑的新时代》,与保罗-沙尔合著

高级研究员安雅-卡斯珀森(Anja Kaspersen)与新美国安全中心(Center for a New American Security)的保罗-沙雷(Paul Scharre)就技术与战争交叉领域的新问题进行了对话。

2024年7月2 日• 播客

控制论、数字监控和工会在技术治理中的作用,与Elisabet Haugsbø合著

高级研究员 Anja Kaspersen 与技术联盟 Tekna 主席 Elisabet Haugsbø 就她的工程之旅、AI 时代的弹性等进行了交谈。