Yahoo! and YouTube: Balancing Human Rights and Business

JULIA KENNEDY: Today we're going to get a little bit of background about human rights and business, and some of the issues that are going on, from our wonderful representatives from NGOs [non-governmental organizations] in the room.

Then we're going to run a few mini-cases from Yahoo! and YouTube, as well, later in the program.

Our first speaker is Rachel Davis. She's an attorney and an adviser to John Ruggie. And she also helped write the Guiding Principles on Human Rights and Business, which was adopted by the UN Human Rights Council over the summer of 2011. Rachel is taking a lead role in creating a new NGO called Shift, which will help companies implement the new Guiding Principles.

I've asked her to give some background about the Guiding Principle, ways they can help businesses address dilemmas around human rights, and some of the challenges of implementation.

RACHEL DAVIS: I want to do two things very quickly. The first thing is just sketch some background for you of why did the UN get involved, and what is the significance of these Guiding Principles, now that we have them.

The second is to talk briefly about some of the implementation issues we are seeing through the work that we're going to be doing with Shift on a cross-sectoral and cross-regional basis. I know the focus of today is ICT [information and communications technology] companies, but this is just to put it in a bit of a broader context.

So, a little bit of history. Previous efforts at the UN to address this issue really—and by "this issue," I mean setting some authoritative standards for the business and human rights debate—really got stalled and broke down in deep divisions, between the business community on one side, civil society on the other side, with states largely absent from the room.

In 2005, the then-Secretary General, Kofi Annan, appointed John Ruggie, as Julia mentioned—John is still a professor at Harvard Kennedy School—to set out on a process of consultation, research, exhaustive and exhausting, on stakeholder interaction on all continents, to come up with some standards that everybody could agree on.

So that's what we worked on for the best part of the last six years, with a lot of input from companies, as well as from governments and other stakeholders.

In June of this year, the Guiding Principles were endorsed unanimously by the UN Human Rights Council, which is very significant, because we now have a common authoritative global reference point for this discussion.

These Guiding Principles do three main things:

• They clarify what states should be doing when it comes to protecting against human rights abuses by third parties, including business, by setting clear expectations for business, and implementing clear rules of the road.
  
  
• The second thing that they do is they really clarify for companies what is now a global expectation, that companies respect human rights. And they provide companies with a blueprint for what we like to call "knowing and showing," how you are actually respecting human rights, in fact.
  
  
• The third thing that the Guiding Principles do, is that they emphasize the importance of effective remedial mechanisms when things do go wrong, because they do. And they give all stakeholders tools to really test company claims, when they say that "we are respecting human rights."
  
  

So, this is now a common global reference point, and we are seeing it picked up in the standards developed by the OECD [Organisation for Economic Co-operation and Development], by the International Finance Corporation, by the International Organization for Standardization, as well as by voluntary multi-stakeholder initiatives, by individual companies, by governments, by regional bodies, et cetera.

A couple of points on implementation:

What the Guiding Principles really try to do, when it comes to businesses, is set out some clear management processes that businesses should have in place to prevent, mitigate, and address the risk of negative human rights impacts.

This means that there are a few challenging issues for companies to work through. As I said, the Guiding Principles were informed by actual company practice and experience, like those of the companies at the table, in trying to get this right.

The three issues I would highlight are:

The first one: It's really about risk to the impacted individual; it's about risk to the rights holder. So it goes beyond classic—and I think pretty outdated now—conceptions of just looking at the risk to the business itself. This is, though, a challenge. This involves stakeholder engagement. It involves really understanding the perspective of those it will impact upon. So this is new territory for many companies.

The second challenging area of implementation we are seeing, is around the fact that the Guiding Principles apply not just to your own activities as a company, but also to your business relationships as well. So how do you deal with your value chain?

Now, here there are no simple answers. I think people are aware of that. But what the Guiding Principles try to do is provide, if you like, a structured decision-making matrix. So here are the key factors that you as a company need to take into account when you are thinking about:

• What do I do that has been linked to something far down the supply chain?
  
• What are the kinds of tools?
  
• What kind of leverage do I have?
  
• Can I increase my leverage?
  
• How necessary is the relationship? What are the other adverse consequences of termination?
  
  

All these kinds of things. It's about a really nuanced and context-specific approach to decision-making. But it's about having good processes in place up front.

That also applies to the third implementation challenge that we are seeing, which is around, if you like, what we very euphemistically, in UN style, call "issues of context" in the Guiding Principles. What that really means is a situation where you have a direct conflict between national law and standards, and international human rights standards. And you, as a company, are trying to operate in that national jurisdiction, and meet the international expectations, and your responsibility to respect.

A lot of what we're talking about today gets into that. But I just want to say that's obviously not something that's unique to the ICT industry.

A final point on implementation. The Guiding Principles rely on companies putting these management processes in place, through a system of cross-functional collaboration. So they're about dealing with your internal stakeholders, as much as dealing with your external stakeholders. Again, that can pose its own challenges.

But ultimately, by setting out clearly the responsibilities of government, the responsibilities of companies, and the expectation that all stakeholders can now hold both those parties to, the Guiding Principles are really intended to help address collective action problems. So that you get these different parties coming to the table, particularly when you're talking about dilemma situations, to work together and say, "What are our respective responsibilities and how can we find solutions?"

JULIA KENNEDY: One thing I forgot to mention about Rachel is she is a Carnegie New Leader, here at the Carnegie Council. So we're so glad to have her here in multiple capacities.

Next up we have Susan Morgan, who is executive director of the Global Network Initiative, and is working very closely with a lot of the issues we'll be talking about throughout the session today. She's got enormous energy and drive and is bopping around the world constantly, speaking at different conferences, and working with different participants in the Global Network Initiative to make sure that everyone is on the same page.

She has brought enormous information communication technology companies together to agree on these principals on human rights, including Google, Yahoo!, and Microsoft.

She'll talk about the Global Network Initiative and lessons around human rights that the tech sector is learning.

SUSAN MORGAN: I will talk about the Global Network Initiative, why we came about, how we emerged, and then I'll say a little bit about what we do as well. The mission of GNI is really to protect the freedom of expression and privacy rights of users around the world.

We're a multi-stakeholder initiative. So we have companies, human rights organizations, investors, and academics who are active members and participants of GNI.

Really, what we're trying to do, is to address a series of issues that, I think, started to emerge probably six, seven years ago. The issues I briefly describe in this way.

Firstly, there's an exponential growth of the number of people around the world who are using information communication technology. So we have something like 650 million users of Facebook, about five billion people have cell phones, about two billion people connected to the Internet. So there's a huge number of people who are starting to use this technology. And using the technology in a way that's increasingly integral into their social, economic, and political lives.

What we're starting to see, as a result of that, is a lot of government interest around the world, in terms of the way in which these technologies are being used, and the potential impact that they can have. So we're starting to see companies increasingly getting requests from governments—either to share information about users or to censor information that's on the Internet—that might be restricting the freedom of expression and privacy rights of users.

The consequence of that is that the business decisions that companies are making in this sector are becoming really important. So, where data is stored, where servers are located, on what conditions companies will share information with governments—these kinds of things really can impact on the rights of a lot of people in many, many different places in the world. Those are the kind of issues that GNI was established to address.

The idea of having companies working with human rights groups and with academics and investors is not a new thing for GNI. There are other sectors and other industries that have really tried to create similar kind of models in this intersection between business and human rights.

So you can look at the extractives industry; you have something called the Extractives Industry Transparency Initiative. There's a number of things in the apparel sector, the Fair Labor Association, the Ethical Trading Initiative. So GNI is very much modeled on these types of initiatives that emerged some time ago in other sectors.

In terms of the work that we do, the first thing that GNI did was to develop, using these four stakeholder groups, a set of principles and guidelines to guide company action when they are facing requests from governments that might impact on the rights of users.

Those guidelines are based on international human rights standards, and were actually developed very much alongside the UN framework that Rachel was talking about. We've had, for a number of years as an observer, someone in Rachel's team working alongside us. So our principles and guidelines very much mirror the kind of protect-and-respect framework.

The principles cover very, very broad things around how can companies really operationalize into the way in which they do business, thinking through the implications around freedom of expression and privacy. So that's things like the technology that companies are developing, products and services that they are selling, and the markets that they operate in.

So how, as a way of doing business, can companies think about those things?

The principles then look at some very specific recommendations around how to deal with requests from governments about censoring information, or looking at handing user information over to governments.

Alongside the principles and the guidelines, organizations who join GNI—and we're a member organization—all organizations commit to really support and advocate the principles.

Companies take on an additional step, which is really to implement the principles and the guidelines within their organizations.

And then there are a number of other activities that we do.

Firstly, we have a process of independent assessment on the way in which companies are implementing the principles. So we want to be able to establish credibility, both for the work of GNI, and also for the actions that companies are taking in living up to that public commitment.

And then, what we are finding now as we develop, is that we are able to use the principles and the guidelines as a basis for policy work and policy engagement with governments, and also as a way of providing an opportunity for companies to work with human rights groups, investors, and academics in a kind of safe space to work on the issues, because I think there are very new issues that are developing very quickly.

There are so many examples in the last 12 months of what has happened, from issues in the Middle East and North Africa to other places—there are many, many companies facing many issues. So we really try to provide a space for organizations to come together and to work through the issues.

JULIA KENNEDY:Let's dive into the cases portion.First up we have Ebele Okobi-Harris, who has a wonderful résumé. She's got a J.D. from Columbia Law School here in New York and an MBA from HEC Paris [École des Hautes Études Commerciales de Paris], which was ranked by the Financial Times as the number one business school in Europe.

So she has this diverse career in law, consulting, and now at Yahoo!, and has worked both in the private and NGO sectors.

But the single thread throughout that career, as she said, is to do work with impact. I learned about that through a compelling BusinessWeek online column that she has written.

At Yahoo! she combines this legal, business, and do-good impulse as director of the business and human rights team. Before we jump right into the cases, why don't I ask you what it's like to be working in human rights at Yahoo!.

EBELE OKOBI-HARRIS: It's unique. Yahoo!'s Business and Human Rights Program is the only program of its kind, certainly in the information and communications technology sector and, I would argue, in a lot of other companies.

I think for my part—in doing a lot of CSR [corporate social responsibility] work, you often have to make the case of why it's necessary. I feel that I'm lucky, because the case was made before I was hired. I think Yahoo! learned a lot of very difficult lessons in some of our adventures in other countries. So I feel that I'm quite lucky because I work for a company who takes it incredibly seriously, from senior leadership on down.

I think the other thing that I would note is that we are in a position to actually help the company make strategic decisions. I think sometimes CSR at companies can be about marketing, which is all fine, or about PR. I think at our company it's very specifically placed within legal to serve as an opportunity to have a direct impact on business decisions before and as they're being made.

So I think it's a privilege and it's a unique opportunity, and it's the best job I've ever had.

JULIA KENNEDY: Let's talk about a specific dilemma that you've dealt with in your time at Yahoo! Tell me about a time that Yahoo! made a judgment call about content posted to its photo site Flickr, which is owned by Yahoo!

EBELE OKOBI-HARRIS: I'm going to try to give you a basic landscape.

I was tipped off by something that Rachel Davis said when she talked about the tension between national law and international human rights standards. I think this makes it very difficult—and you'll hear from my counterpart at YouTube—from classic CSR problems, where you're talking about differing standards, for example, like child labor. Or if you're talking about the extractive industry.

In this case, you're talking about national laws that have clear sanctions. So it's not a matter of, do you meet a minimum standard that's different in, say, Thailand, than it is in the United States, when we're talking about child labor? You're talking about a law. So that's one.

And two, if we're talking about privacy and free expression, there is absolutely no country in the world that allows for a completely uncensored Internet. So, a lot of times, when we're thinking about this as a problem, they say, "Well, companies, you should never censor."

First of all, if it's a law in a country, you are bound to follow that law. Two, no company is in a position where they can't censor, because every single country in the world has a censorship regime of some sort. The joke that I always make is —censorship, to people, is always what they don't agree with.

So an American would say, "Of course child pornography shouldn't be online." We would all agree with that. That's something that you can't even argue with. But that in itself is a government saying, "This we do not want."

The last centering point I would make, is that there is also no government in the world that has a sophisticated police force that is not interested in solving crimes through getting user data. So, we focus on China, we focus on the Middle East.

But every single country in the world is interested. And if you look at Google's map of government requests, you'll actually see that the country that sends out the most is the United States. That's just something that grounds this in what the reality of it is for companies.

Having said that, there are a lot of issues that I've dealt with in my role. I think one of the most interesting—given everything that's going on in the Middle East—is the issue with Yahoo!'s property called Flickr. It's a photo-sharing site.

As you all know, Egypt is one of the beginning countries of the Arab Spring. Right now, we feel a huge amount of enthusiasm for what the people have managed to do. There are a lot of articles about how people were using social media to basically change the world—not that social media did it themselves, but they were using it as a tool.

Flickr, as I said, is a photo-sharing site. There is a very well-known activist in Egypt. He goes by the name el-Habaway, but his name is el-Hamalawy. He has been a long-time Flickr user and he uses Flickr to post a variety of photographs that he has taken.

He also is a member of a Flickr group called Piggipedia. Piggipedia is a group that was started years ago to identify members of Egypt's secret police force. It was done because they wanted to make sure that these people were people who were brought to justice.

In the middle of the uprisings in February—this was the time when Egyptians stormed the secret security headquarters, and they ransacked it, and they were able to get disks—El-Hamalawy took the disks. And he posted images from the disks onto his site. And below that he had a caption that said: "These are secret police people. They must be brought to justice. This is who they are," et cetera.

JULIA KENNEDY: We're in February of 2011, beginning of the Arab Spring. You're at Flickr. You see this post. How do you handle it?

Take a look at these pink Flickr Community Guidelines [audience was handed an excerpt from the Guidelines]. See if that gives you any insight on how to handle this post.

What first comes to mind if you hear that this photo has been posted? What are some of the questions you want to ask about it or what's something you want to address if this has been raised, attention about this photo has been raised?

PARTICIPANT: It sounds a little like Law and Order ripped from the headlines, because they're more or less saying "You better do something to these people." It definitely conflicts with "do play nice" [part of Flickr guidelines]. That strikes me right away as what you're dealing with. I'd say you'd have to take it down.

JULIA KENNEDY: What do others think? Anyone want to defend leaving it up?

PARTICIPANT: Was that decision made?

JULIA KENNEDY: We're playing it out. We'll let you know.

PARTICIPANT: Not necessarily to defend it, but the guidelines seem a little vague, and judgment calls need to be made. So you could argue either way, whether you take it off or leave it up there, that you're adhering to or not adhering to some of the community guidelines. It feels like maybe they need to be defined a little more strictly for any real decision to be made based on the guidelines.

JULIA KENNEDY: Any other thoughts?

PARTICIPANT: There's clearly a theme with Flickr. You didn't take the photos. If they're not your photos, that you mustn't upload them. So that seems like that's a large part. There's no misunderstanding that he knew that they were not his photos. And he thought it was appropriate to use the Flickr format to promote photos that were definitely taken by someone else.

JULIA KENNEDY: That is a key element in the decision-making process at Yahoo!. Do you want to take us to the next step in this, and tell us what you were considering?

EBELE OKOBI-HARRIS: I'll start by saying that one of the reasons I picked this is because this is a workshop. And I wanted to work through some things that worked, and some things that maybe didn't work so well.

We have a Business and Human Rights Program. And we are supposed to be contacted whenever there are these types of issues. To your point about vagueness, the Business and Human Rights Program has always been focused on government requests for data.

So when Flickr saw this, they thought, This is not a government request, because it came in through a Flickr community member. Actually more than one clicked the "abuse" button and said, "This is wrong. These should not be up here."

So the Flickr community managers didn't escalate it because they thought, "Not a government request."

They went through the entire decision process without telling me. I actually found out through two different GNI members the value of being in a multi-stakeholder initiative.

Because people, nonprofits, instead of immediately going out to the public and saying, "Oh my god, Yahoo! is a horrible company"—although some people did say that, and that's fair—instead of doing that, they came to me directly and said, "You know what? I know that there's someone here who focuses on this. What happened?"

So the decision that was made by Flickr is that: Look, there are a number of issues with this photo. Number one, it's basically threatening someone. And also, it's not their own photograph. So the decision was made to take it down.

You pointed out one of the things to think about. The community guidelines are vague. One of the things I would posit: there are hundreds of millions of pictures posted on Flickr. To figure out a guideline that would address every single layer of ingenuity of someone who's interested in posting pictures would be completely impossible.

So one of the reasons that the guidelines are deliberately left open, is because it's impossible to cover everything.

And the more detailed you are, the more someone can argue and say, "Ha ha, you have not covered this; that means that it is allowed."

So there is an aspect of judgment call being made.

Anyway, to answer your question, the decision was made to take it down. But the public reason given was because those pictures were not his. And not only were they not his, he explicitly said, "Hey, I stole these—"

The caption said, "I took these from the security office which we have just ransacked. These are not my pictures. Here they are. Bring this person to justice."

PARTICIPANT: Based on what you said, do you feel that these guidelines could be, or have they been, strengthened to get to a higher level of principle?

Because when I look at them, I can see how they would apply to my Saturday soccer team. And that I don't want to put a nasty comment on someone else's soccer team. But it feels that they could be up-leveled to a point where they speak to some of these principles. Are you working on it, or have you learned from those experiences?

EBELE OKOBI-HARRIS: That's an interesting question. Yes, I think we've definitely learned from the experience.

I will say this. And this is I think what gets to the tension. Flickr wasn't necessarily created for this particular instance. In fact, the majority of people who use Flickr use it to do stuff like post their grandkids' pictures, and that sort of thing.

I think you point out one of the tensions, and one of the opportunities for a company, when you realize how people are using your product. So, to your point, the guidelines are written like this, because that's what the vast majority of people use Flickr for.

But now there is certainly an opportunity for us to think about addressing a different audience.

It is for this reason that we are having, at the end of October, a summit where we're bringing together activists—we're also bringing together people from YouTube—to talk about this specific issue.

So how can companies figure out user guidelines that are flexible enough to cover how most people are using these tools but are also specific enough in some cases where they need to be?

I think YouTube has done an interesting job with some of their videos, and you'll hear Abbi talk about it. But how do we create ones that are specific enough to cover very specific instances that do not represent how most people are using their property?

PARTICIPANT: I actually just wanted to expand, and give you perhaps a further hypothetical.

You say that the public reason for removing the pictures was because he made it very clear they were not his.

If, in this case, let's say he got a file, had the names of the secret police, and managed to go track them down himself, and take his own pictures of the secret police. And still put the same message—"These are secret police; I've identified them; they need to be brought to justice." You can claim that he's putting those people in direct harm.

Would that have been sufficient? Or was it only because it was not his pictures that allowed you to remove the pictures?

JULIA KENNEDY: That's a great question. Why don't we have people discuss it.

PARTICIPANT: I would refer to "don't be creepy" [part of Flickr guidelines]. That could be the type of thing where—"creepy" is probably a term that is used to refer to other behavior. And maybe that's again intentionally vague for the community to report on what's creepy.

But where YouTube is clearer about stalking and so forth, "creepy" encompasses a lot of things—threats, et cetera.

JULIA KENNEDY: He might say, "Well, this is free expression. I am a political activist and this is free expression." I don't know, but that might be a response.

PARTICIPANT: I think one of the things we're talking about here, and haven't brought up on these Flickr community guidelines issue, are values.

Most companies, if not all, have a value statement and mission statement. Possibly—these are just dos and don'ts—but a level up from that is what is the overall values and mission of Flickr within Yahoo!

I don't know Yahoo!'s mission statement, the way everyone knows Google's, which is "do no harm."

But if there was a values or a mission statement that was more specific, like "create positive change" or something like that—if we're challenging our companies to create a difference in the world and then putting that in our values or mission statements for those businesses—then we can also argue that the impacts of these are both positive and negative.

It doesn't make it any easier to solve this, but in many respects I consider these just very legal. What to do/what not to do is not the same thing as looking at your business, seeing the impact it makes in the world, and saying, "What are our values and how do these actions live up to these values or not?"

JULIA KENNEDY: What we have posted here is just a short excerpt of much of the material you'll find on Flickr having to do with values and mission statement. So there's a lot more on the website. I just wanted to give out a short something for the case.

EBELE OKOBI-HARRIS: I thought it was interesting you said that Google's thing is, "Do no harm," because that actually isn't it. [It's "Don't be evil."]

But having said that, to the point, I think there are a lot of overall mission statements. And that's great. But that does not help you with this specific instance.

So, for example, if we talk about wanting to be a platform where people can come, engage, and share information, that does not help you with, "Is it okay for someone to post a picture that explicitly intimidates or harasses another person?"

Because you could make a human rights argument either way. So, for example, what if someone saw that and said, "Ha ha, I know who this person is, and I will go and kill that person"? Are you as a company responsible for that behavior? Should you be responsible for that behavior? Even if that person has done all the horrible things this person has said, where do you stand as a company?

I say that, only to say that it is wonderful to have larger overarching statements. But what we're talking about here—which is critical for all companies to think about—is when you get in the weeds, what guides your decision-making when you're there? And understanding the overall impact of what those decisions could be.

So if we left that up and said, "Okay, we don't think there's anything wrong with it," and something happened, would we then as a company be responsible for it? I think this is what makes it really complicated decision-making.

PARTICIPANT: I think that's a great point. Because one of the things I know from working in tech for a long time, is that there's always going to be a new technology that you didn't anticipate. Government, law, and regulation, or even standards, cannot keep up with that. Technology is going to move faster than government.

So I think one of the really interesting things that you're raising is that you are inviting a dialogue about this, because there are many places to post photos on the Internet. So you could put them on Flickr, you could put them on any number of sites.

The conversation that you're inviting is about what is appropriate use overall of media. And how do we as a global community have a conversation about this, because it will happen, whether it's on Flickr or somewhere else.

PARTICIPANT: I have a question about the overall process, in terms of how you manage these types of things, and how quickly you take them down. Is there a process in which you can take them down while you're viewing them? These things are so viral that if a photo is up there for 30 minutes, it can really reach a larger audience?

If it could cause harm, how do you react to that? Do you take it down and say, "This is under review"? Or do you take the time to review it before making that decision?

EBELE OKOBI-HARRIS: I'll start by saying, every different property at Yahoo! has different people who are responsible for taking it down. So Flickr's capabilities are different from, say, Yahoo! Front Page.

Flickr does have a way of hiding content without taking it down. Also, as you can see from the guidelines, there are different levels of takedown. For example, there are some things where if you put that up, they will close your account immediately.

And there's absolutely no right of appeal. There are some things they might hide, contact the person who put it up, and say, "Hey, look, your account very broadly is violating community guidelines. You may want to look at it and take something down."

So it depends on exactly what the problem is with the account.

But the short answer to your question is yes, there are different ways of doing account or issue takedowns.

PARTICIPANT: I wonder if you could describe, from a strategic perspective, how the decisions are made—fundamental decisions on how the platform is built.

So for instance, do you ever discuss at Yahoo! "What type of platform do we want to be? Do we want to be the most ethical? Do we want to be the most open? Do we want to be the biggest?"

Those very fundamental tradeoffs that really are at the heart of the matter—then what you see down the road in a way are the result of those decisions.

But if they aren't made intentionally early on, it's hard then to fix it when it happens. So, can you describe the process through which you and your group engage in those conversations?

EBELE OKOBI-HARRIS: In terms of what Yahoo! wants to be as a company, those are conversations that are ongoing. Not just from a business and human rights perspective, but just from a company perspective. I think certainly business and human rights should inform that thinking.

But it's not the only question. There are a lot of other questions that go into a company deciding that.

Having said that, I think one of the reasons Yahoo! created this department, is because they are very much interested in getting input from a human rights model, like, "What should we be thinking about? So as we're thinking about who we are as a company, what are human rights issues we should be taking into consideration?" That's one of the reasons this department was created, so that business decisions could be informed with human rights issues.

SUSAN MORGAN: Ebele, you said that one of the issues was that this didn't actually get escalated as you would hope it would have. That's something we see very often in these kinds of situations.

It's really important that a sufficiently senior level of decision-making authority comes in early enough to work out what the problem is and to work out what your public response will be when you get asked. Would this be dealt with differently now?

EBELE OKOBI-HARRIS: Yes. Certainly it would be escalated. As I said before, because the Business and Human Rights Program is understood as,"Oh yes, if we get a government request"—so in this case, like I said previously, people said, "Oh, not a government request, so this is not an issue."

Now the understanding of what should get escalated is much broader. So people realize that anything that might have a potential human rights impact, that might have an impact on freedom of expression, we should escalate that. I think that would certainly be different. So that would be escalated. It would be discussed at a much higher level than at the community manager level.

JULIA KENNEDY: I'm curious. You had mentioned there are issues of internal implementation, as well as external implementation. So, say, you are at Yahoo!—and I'm going to turn this to you next—but, first, for the participants. Say you decide, Okay, this photo needs to come down off of Flickr, and that decision has been made.

Then, in terms of implementing it, you get pushback from an employee. And you get an email from a user, maybe another activist, saying, "Hey, this photo is important, it's part of free expression"—then how do you respond to that challenge? I guess that is the question.

PARTICIPANT: I guess I'm wondering if—rather than trying to fit all the different objectives of the different people who are using file sharing and photo sharing—maybe the answer will be:

"Look, it just doesn't fit on Flickr. But you know what? Six weeks from now, we're going to be rolling out a new product where you can share photographs, which will have a different set of guidelines and a different set of objectives. Just be patient. This is where we're headed."

Could that be a possible answer?

JULIA KENNEDY: So maybe a new different type of community.

PARTICIPANT: I want to echo what an earlier questioner asked. If I understood what you were saying, it sounded like you find it difficult to have principles-based lawmaking as opposed to rules-based lawmaking.

EBELE OKOBI-HARRIS: No, no, no, no.

PARTICIPANT: These guidelines are very specific rules-based. We're trying to understand. We would have to read this many times to understand the culture behind Yahoo!, and what you're trying to achieve. Whereas if you had a principles-based rule-making, it would be much clearer for people, much easier for you.

Perhaps with new products, if you had a larger ethical umbrella under which everyone knew how your company operated?

EBELE OKOBI-HARRIS: I just want to address that really quickly. When I was talking about the rules, I'm not saying we don't have principles by which Yahoo! is operating. I'm saying—for purposes certainly of this conversation—if people need to understand exactly what can I post, and what can I not post, they actually need to refer to rules.

And I was also saying that even if you have principles, it still doesn't answer the basic question. So, for example, if your principle is, Yahoo! says "We're the premier digital media organization, and Flickr is all about people being able to share information and share photos with each other"— that doesn't answer the specific question. There are always going to be new and emerging issues.

It doesn't answer the specific question. The joke I always make is—I don't know how many of you are familiar with Monty Python—"Nobody expects the Spanish Inquisition." Nobody expected the Egyptian revolution. You can't anticipate every single potential wrinkle that is going to come.

Even if you say, "Yes, we want to be an open platform"—open for what, open to whom? That's why I started by saying there is absolutely no way that a company can have a completely open platform.

So yes, I think it's very important to always be referring back to what your principles are, which are that we want as many people as possible to use our platform. We want people to be able to use it to photo share.

The other point I would make, is that what Flickr started as is actually very different from how this user wanted to use it. That's one of the reasons why we want to continue to have this conversation, because Flickr is actually not supposed to be a place that you post pictures that someone else took, so that that person can then be arrested. Flickr is supposed to be a place that you can post pictures that you find pretty or engaging and whatever, and people comment on them. It's supposed to be about people who are photographers, coming together to share things.

Now, that's not to say that Flickr can't emerge or evolve. But Flickr's original founding principle is actually completely contrary to how human rights activists want to use it. Now again, that doesn't mean that we can't evolve. And that's part of the reason why we want to have these conversations. But this is actually exactly the opposite of what Flickr meant to be and what Flickr was founded to be.

So to your point about principles, there were principles. And this was actually against them.

PARTICIPANT: It would be much easier to have those principles in a place where you could read them and understand them.

EBELE OKOBI-HARRIS: They are up there.

JULIA KENNEDY: They're up there. This is not everything on Flickr. I've gotten requests for more information and I've just been responding because there is so much out there.

PARTICIPANT: One of the interesting elements of this example is this notion between who the request was coming from—if it was coming from an individual or a government.

To pick on the Egypt case—but it's quite a relevant one since the Arab Spring, and perhaps will continue to be so—who is government and who is the public?

The person who posted the objectionable images a few weeks from now might be a member of government, and could make a claim that these pictures are the property of the state, or the property of the people, et cetera.

I'm not arguing that this would actually change the decision. It's more of a question of how did those judgment calls get made. I guess it's again going back to the operating principles—not the values principle, but the operating principles—where you at the end of the day still refer to the guidelines.

EBELE OKOBI-HARRIS: I don't think I totally understand.

PARTICIPANT: In part, I found it an interesting scenario in that the request might have been coming from an individual. Or it might have been coming from the government. But because the situation was so hazy, those lines are quite blurred and might continue to be so, as uprisings and revolutions continue throughout the world. That's just an aside.

But I guess, perhaps beating a dead horse of values and guidelines—but, indeed, because you can't anticipate every single scenario, what is in the end the operating principle? Is it keep the site as open as possible as long as these rules are obeyed? Is that what it basically boils down to?

EBELE OKOBI-HARRIS: No. That's why I'm saying that I think this, for us, was a very interesting thing, because Flickr is not a photo-hosting site. Flickr has been very explicit about saying, "We don't want every single photo that you might want to post. We don't want photos that you haven't taken. We don't want a whole bunch of pictures, photos that you have snagged from other sites. Or because you're a huge Lady Gaga fan and you've put together a collection."

Flickr's values or principles are that it wants to be a site where people who really love photography can share their pictures, can get to be better photographers. That's what Flickr's principles are.

It's not, "We want it to be as open as possible." That's what I go back to. So yes, that's the overarching principle. But we're realizing that people are trying to use Flickr in different ways.

But it's a company decision—and it's separate from human rights—to say: Well, do you want to be a place where people post any photo at all? Or, do you really want to focus on being a site for people who are photographers, who are really interested in getting to be better? Of, for people who can share comments like "Look, I took a picture of my grandchild"? What do you want to be?

So again, this is the principle, and the principle I think was actually quite clear to Flickr users. But I think it's an opportunity for us as a company to figure out if people are using it in a different way. And, if there's a human rights impact, how can we have a conversation with human rights activists, and see how we can use it differently? That goes back to the question here.

So that was one of their suggestions. And I think YouTube—and Abbi will talk about it—has done a really interesting job of saying that there are different rules, or they will review images or video that have a human rights aim differently. I think that there are a lot of questions that go along with that.

One of the questions that I asked, when that was recommended is: For us, how do you define a media site? How does that get defined? So, if Flickr says, "We are not a photo-sharing site," is that really different just for human rights activists?

If that is the case, then how do we define who is a human rights activist? Anyway, there are a lot of really interesting questions which I think Abbi will get into, because I think YouTube has done interesting work figuring out what that means.

But for us at Yahoo, we had not reached that point, because the decision was made on two bases. One, that it wasn't theirs, and Flickr is quite clear about it not being yours. And that the actual caption on it was threatening.

JULIA KENNEDY: I'm going to fast-forward a little bit because we do have a lot of great stuff to talk about from YouTube.

I do just want to cap the story by saying that Yahoo! took down the photo. But they've allowed him to continue posting other photos. So, if you go onto Flickr and look up el-Hamalawy, you can see he's still posting actively on the site.

EBELE OKOBI-HARRIS: And Piggipedia is still up.

I guess the only thing I would say, is that this is an ongoing conversation at Yahoo! and at Flickr, and that through GNI, but also separately, we have invited a further conversation with human rights activists. So we're sitting down, having a workshop, and saying, "Specifically, how can this work? What are the details of it?"

Because, yes, we can talk about overaching principle. But how does it work at Flickr? And what is it that human rights activists are looking for? And how can we evolve and engage and respond to those specific requests?

JULIA KENNEDY: Thank you so much. That was very insightful and very helpful.

Let's move one now to talk about YouTube. Abbi Tatton is here. She spent a decade reporting on the Internet and politics for CNN. So she has been thinking about these issues for quite a while. She joined YouTube as Manager of Global Communications in 2010.

She has brought along two very compelling videos to share with us. I just want to give the warning to our webcast viewers, as well as you here in the room. Some viewers may find the content disturbing in these videos. If you prefer, we'll let you know ahead of time and you can cover your eyes, leave the room, or whatever. But they're also very compelling, so we thought it was important to include them.

Before we jump into the cases, Abbi, why don't you tell me, also, what it's like to be working in these issues of human rights and communications at YouTube, and how the challenges you face are unique or more universal?

ABBI TATTON: Well, I think the first thing to start with with YouTube is scale. And to also note that—just listening to Ebele talking about this—it's not unique. We are all facing some similar and very new challenges.

YouTube is six years old. It was started with the motto "Broadcast Yourself." I don't think anyone could have anticipated how much everybody would. We're now at 48 hours of content uploaded every single minute. So by the time we're done here, thousands and thousands more hours of content will be on the site that none of us could possibly sit through and watch, three billion views a day.

So with the motto "Broadcast Yourself," this is a platform for free expression, giving everybody a voice. And we believe that that's a great benefit to society, access to more information.

But, at the same time, what we're trying to balance is a platform for free expression, which necessarily has to have rules. And what those rules should be. And how we strike the balance between being this place where everyone has a voice, a place for the free exchange of ideas, and a platform that's safe for users as well.

Sometimes this is really simple. If content is illegal—if content, for example, contains child pornography, we will omit it. Again, as Ebele was saying, all around the world the jurisdictions might have slightly different interpretations of what that illegal content is. We don't. We have a zero tolerance policy on that. We think that's pretty simple to enforce and to implement.

On other things, as we'll look in a couple of case studies, it's really not that easy. You'll see our community guidelines. Again, we've spelled them out. I heard the comment that it seems vague. It seems vague from Flickr as well. But, just as we were hearing already, when you have this much content, how can you write every single policy to come up with every new challenge that we couldn't possibly anticipate yet?

I think looking at a couple of videos will illustrate this, just to say that these are tough calls. We review video carefully. And we scratch our heads, sit around, debate, and talk about it. I'm not saying we always make the right decision, but we look at this content very carefully.

JULIA KENNEDY: Great. So let's look at the first one. This is an excerpt from the first video we're going to ask you consider, if you could put yourself in the role of a reviewer at YouTube, and decide whether you think this should stay up or not. As you all know, the guidelines are published.

[Video presentation: excerpt from "How to Make an Exit Bag." Warning: this is a very disturbing video about how to commit suicide.]

ABBI TATTON: This is about four and a half minutes long. I'll stop it there.

JULIA KENNEDY: Put yourself in the place of a reviewer. What stands out to you about this video, and does it raise any concerns for you dealing with human rights?

PARTICIPANT: I don't see why a company is required to post that. I'm not saying that it can't be on the Internet. But it would be up to you whether you want to give people ways in which to kill themselves. I don't see where you'd be violating anyone's right of free expression as a company.

ABBI TATTON: By taking it down?

PARTICIPANT: Yes.

JULIA KENNEDY: Obviously this is very disturbing content, and these are questions to mull over. But I'm wondering if anyone has any other perspectives on it.

PARTICIPANT: I think those are issues that—there are professionals who deal with issues like that all the time who are far better—they've thought about this in a way that maybe we don't always consider. Doctors have principles of "first do no harm." I'm wondering if there might not be some guidance among such professionals on how to even think about an issue like that.

JULIA KENNEDY: And so how would you say that should inform the reviewer at YouTube when considering whether—

PARTICIPANT: In other words, you wouldn't necessarily want a reviewer to make a decision purely based on a gut feeling. But, if there was a set of issues that had been debated by professionals, like doctors. Or people who are dealing with problems that the aged have, when they're facing a short lifespan with tremendous pain, and there's no hope for improvement. There are such complex issues around that—

JULIA KENNEDY: So, you would call in the support of, say, a medical ethics board or something, and call in a doctor to see what they think?

PARTICIPANT: Yes. At least to create principles around it.

JULIA KENNEDY: Okay, great.

PARTICIPANT: I think, number one, this is a piece of a video. We don't have the full story. I think it's important that this information is available. I'm a health professional myself, so I can speak on that side of things. Whether it's appropriate that it be available on YouTube, rather than—as the gentleman down here was saying—somewhere else where there is a professional grounding to that information, I don't know.

But this is open information and it can be found from many, many locations.

PARTICIPANT: I guess there are absolute answers, and nuanced answers. In this case, it strikes me—just as a point of view—that there has to be a nuanced answer. This could be seen as public health information. There are many ways to look at it.

I think the issue is that the platform is an open platform. That you don't pay to view this. So the access is so easy, so immediate, and so global that, in a way, it is a company's responsibility to decide either in absolute terms, "This is a yes or no."

And, once you have made that decision, if it's a yes, then to decide how it will be further filtered. So that it's either a separate community, for example. And then, even within that community—imagine that this could be a medical library, for instance—you would still choose which documents are worth being showcased in that library. So there are a number of criteria.

I guess the complexity for you, I'm sure, is volume, just sheer abundance of information.

So I wonder if you could describe how do you deal with these nuances. And is it now leading you to reconsider the openness of this platform, and the free access to the information? What is the debate within your company now?

ABBI TATTON: I think the first point I'd want to make is what you brought up. Which is—we just saw a snippet—there was no greater context. If this was in a medical library, there might be more supporting information. There might be more information around it; there might be a medical professional introducing it.

This is YouTube. There isn't. There is a video. There is some what we call metadata, which is a horrible term, which just means the tags around the video which show up when you search. So the metadata, what a reviewer would see when they were looking at this video, is just the video.

That's the same thing as a user sees. That user could be a 15-year-old who just searched on the word "suicide," and this video came up. Other tags on this video: "how to," "helium," "exit bag," "suicide," "euthanasia"—all these things. So that is what the reviewer is looking at and taking into account.

The reviewer is also looking at our policies and applying every video according to our policies. When it comes to this—we have a lot of different policies for everything from spam to nudity. But when it comes to this one, you'd be looking at a policy which is promoting a dangerous act that has an inherent risk of physical harm or death, which would come down.

You're also looking at a policy we have, around acts that could be copied by minors, because we have so many young users on the site. So, those are a couple of things.

Shall we go on and describe what happened?

JULIA KENNEDY: Yes. It looks like we have a couple more comments before we do that, if that's okay.

PARTICIPANT: I think it all comes down to the bottom line of a policy judgment call based on ethics, because legally the company has no obligation to respect your right of freedom. The right of freedom is only vis-à-vis the government. Companies can do that.

And then I would think that the company is protected by the statute 47 U.S.C. 230, which means the provider is shielded from publishing these things. So, legally you can. It all comes down to a policy judgment based on ethics.

JULIA KENNEDY: I think we need to move on, but I'd love to take your comments in the next section.

So what steps did YouTube take?

ABBI TATTON: This one was hotly debated. There was no clear answer for any of us on this because, like you said, there's no legal reason that we had to take it down. At the same time, we have young users.

What we ended up looking at and balancing was our principle of access to information, leaving up as much information as possible. At the same time looking at—we have this "harmful and dangerous" policy, which is a video that "promotes doing something harmful and dangerous."

Was that video actually promoting it? That's something we took into account. It wasn't a cult going out telling you to commit suicide. It was giving you information should you be looking for that.

Another thing we were looking at is just the fact that this is just a highly debated political issue. And what would we be doing if we were taking it down from YouTube.

However, what we did do is put in some extra safeguards.

We have a policy where we can what we call, "age gate" a video. That means you can only access it if you are a logged-in user over the age of 18. So that immediately reduces the amount of young people that could quickly access that video.

Another thing we've done around the issue of suicide—because we have so many young users—is people searching on "suicide" search terms will bring up, basically, a house ad for the National Suicide Prevention Hotline. So it takes you to other information.

So yes, you will find this, you can get to it over you're 18. But the first thing you'll see is "call this number if you're looking for help."

Is that a perfect solution? I don't know. But that's where we arrived at, given that we are trying to give people access to information, and keep up as much as we possibly can.

JULIA KENNEDY: One more comment, and then we'll move on to the final case.

PARTICIPANT: Just a question in a situation like this. Obviously—and this is for Flickr, too—how much user input or concern are you looking at in terms of, for lack of a better term, community policing of a lot of the content, given that you probably just can't keep up with how much content you have?

ABBI TATTON: I wanted to take a step back just for a moment and look at what does that look like on our end, with 48 hours of video uploaded a minute? I'm not sitting there watching it, right? It doesn't work that way. We couldn't possibly prescreen all the content that's going up on the site.

So we rely on our users. The reason we put up these community guidelines—so everyone can see them—is because it's our users who are our first line of defense.

Underneath every single video on the site is a flag button and a way to flag that video. You can flag it for a number of different things. Whatever you flag it for, it will go to a reviewer. A flag does not take down a video; it gives it to a reviewer, so the reviewer can then take a look at it.

Dealing with scale, we use technologies to try and review things as quickly as possible. For example, if you are a user, and you just flagged 30 Justin Bieber videos none of which should come down, maybe your 31st flag won't be reviewed quite as quickly as someone else's flag coming in.

So there are algorithms like that that assess the flags coming in.

We have a team of reviewers who are working in different time zones around the clock. So videos are being reviewed all the time, to the point of thousands and thousands and thousands a day.

JULIA KENNEDY: Great. Thank you for those wonderful questions.

I also want to thank you for your bravery in sharing some of these case studies. It's very helpful for us.

ABBI TATTON: I think three of them I suggested you wouldn't even show. You need quite a thick skin to watch them.

JULIA KENNEDY: Yes. Speaking of which, the next one I think, depending on your sensibility, might be even a little bit more hard to watch. With that warning, why don't we show it?

ABBI TATTON: Some context for this video.

Again, if you are a reviewer, you don't know anything beyond the video you're seeing, the tags below it, and any information that the user gave you. Sometimes a video won't be flagged until a couple of years after it was uploaded. Or sometimes a video will be immediately talked about so much that it's being flagged, and it's in everyone's mind.

Going back to the summer of 2009, in June, when we saw a flood of videos coming in from Iran, this was the context. News organizations had been kicked out. News organizations were setting up social media desks to look at the material that they were finding online and assess it. So with that background—

[Video presentation: excerpt from "Iran, Tehran: wounded girl dying in front of camera, Her name was Neda." Warning: this is a very disturbing video.]

JULIA KENNEDY: The name of the young woman featured in the video is Neda. She went into cardiac arrest and died during the 40 seconds of this video. The question is: Is it a violation of Neda's privacy to leave this video up on YouTube? Let's start there.

PARTICIPANT: Going back to the guidelines without any other context, it looks like it's just showing somebody dying. So it's hard to say that it has some other meaning beyond that. You've provided us the context now. But that definitely is not clear from seeing the video. Perhaps given the particular culture, it might not be appropriate to leave a video of somebody dying.

JULIA KENNEDY: Are there any signals of context that you can find? Say this is the page that you are presented with, as a reviewer.

ABBI TATTON: Let me just point out a couple of things.

You have a title which is very straightforward. It's not alarmist in any way. It's documenting a statement about the video. There was also extensive text underneath this particular video describing what happened.

So, a video with absolutely nothing showing someone dying—we have policies against graphic violence, shocking content, all of this. With this additional context, does that change your mind?

JULIA KENNEDY: The title is difficult to read, but it does give context to the video.

PARTICIPANT: Obviously, from looking at your guidelines, it says, "Don't post videos of dead bodies." I think I have seen this video, though, or something like it, on the news.

So I wonder from the angle of this, how does the news affect what goes on YouTube, too? Do you look at content that's posted on CNN, or other broadcast news stations, and look at that, compared to what's on YouTube? And does that have any impact on your policies?

ABBI TATTON: Again, going back to the fact that we have guidelines, and shouldn't we take this down. If this same video was shown elsewhere— say we had taken it down and then CNN did a news report that was exactly the same video with eight words of introduction. Wolf Blitzer saying, "Now we're going to show you a disturbing video from Iran," and then played it; would that change it?

And, if so, should that just be Wolf Blitzer at CNN? Who is the news? Who is allowed? This person who shot this video, is he or she not just as important a news documentarian as an established news network?

PARTICIPANT: Especially, given the rise of blogging, especially given you have a situation in Iran where news reporters were kicked out—

ABBI TATTON: Four days before CNN's last—

PARTICIPANT: So, you wouldn't have a news reporter who is respected, or understood to be one by the Western world, introducing this piece. Then, is it de facto not news, because there is no one from the West witnessing it?

ABBI TATTON: Because no one's waving their credential around, right.

PARTICIPANT: Wasn't this on the cover of Time magazine as a still photograph?

ABBI TATTON: Not on June 20, 2009, which is the point that the reviewer was looking at the video, and deciding whether it should be up or not.

PARTICIPANT: Would it, then, have not been anywhere? In other words, this was a big deal.

The minute I saw that I said, "That was the girl who became the symbol of the revolution at that point."

ABBI TATTON: Right.

PARTICIPANT: So that I don't know why—I mean, it was news from practically the minute it happened.

ABBI TATTON: While we have policies on graphic content, on graphic violence, and on shocking or disgusting content, I can tell you right now that if that same video, which was a full introduction of what the video was, and what happened—had the title, "Wow, watch a girl die, exclamation point"—it would have come down immediately.

We're looking at the intent, what was the intent of this video. That's why we recently did a blog post about context, asking our users to tell us more, to help us out. That's why a video which shows a bloody body being beaten, with no additional context, would be brought down immediately. When we know that that is a human rights activist, and this is the only information that he has to document what happened to him, it's different. And our user will see it differently.

What we did with this video is apply what we call "educational-scientific-documentary-artistic policy." Sorry, that's a mouthful.

But essentially nudity comes down from YouTube; artistic nudity is a whole different case.

Same with this. The documentary value of this video is so clear, it was basically the turning point in this news story. It would be very difficult to take this down.

Another challenge when we're talking about scale; search the word "Neda" in YouTube today. We're not talking about this one video. There are 9,000. So say we took down one. What's the policy then on the other 8,000 plus?

This video was uploaded. It was on Facebook, it was on Twitter. It wasn't just on YouTube; it was all over the place. So this was a way that news spread, and a person with a video camera became the news anchor, the news reporter, that day.

So again it is kept up. It's behind an age gate, so it has a warning interstitial. So, just like Julia told you, this was going to be disturbing, YouTube tells you that, as well, and it's your choice to go on. But some difficult issues certainly around that one.

PARTICIPANT: I think what is very powerful is the effect that YouTube, Yahoo!, and other organizations are having on the world globally as a result of documentary-value videos like this. That, I think, is at the heart of the issue.

Basically what we're saying is that the access to information is changing our world. In the past, we had better communication online that was affecting our world. But now the dramatic effect of the video image is going to have implications that we cannot even foresee at the moment.

People who are potentially going through heart operations can now see on YouTube exactly what they are going to go through. And perhaps they either go into it with full information, or be much better informed about whether they want to go through the process or not, rather than trusting somebody.

I mean we are dealing in a world in which some of the tragedies of the past, like the My Lai massacre, the Pol Pot massacre, the apartheid issues, and the Nanjing massacre—the world will know a lot more about all of these. I think it's a big tribute to what YouTube and Yahoo! are doing in this regard, that you are informing us much more quickly.

And we expect that you will not always exercise the best judgment in taking things on or off. That's the price to be paid for the value that you are creating for society. I would view it from that perspective and say leave it up because these kind of things are changing our world.

ABBI TATTON: Just to that point, we do make mistakes. Just the sheer volume means that that will naturally happen. There are human beings reviewing videos. And sometimes they don't have access to all information. Or, sometimes, something will come down for graphic content, where the person next to them might have had a little bit more context, and understood it differently.

That's why we have an appeals process. So if your material was struck, you can appeal that. It's not just our users. Every user can flag a video. But we also hear about videos from journalists, from human rights organizations, from all different kinds of people, alerting us to this material. Because, like I said, we can't possibly prescreen everything that's going on up there.

PARTICIPANT: I just wanted to comment about the news media. Knowing what the news media has built into—the "good news" media over decade, in terms of editorial reviews, quality, and screening—I wonder if we are at a turning point where we either need to accept that some of that has to go out the window. Because, now, the tools that are at our disposal are such that you just can't build that into the process anymore.

Or—the answer is probably somewhere in between—we are at a point where YouTube recognizes that you are a news media organization. And that some of the practices that have been recognized as best practices in news media organizations have to be much more part of your core fabric.

Could you describe what's happening now in YouTube along these lines?

ABBI TATTON: I think we feel pretty strongly, even though we face some of the challenges about what to show, what not to show, and what kind of warnings to put up there.

We've been developing, like you said. The news media has had decades to do this, and we are developing in a very short time frame.

At the same time, we are not a news media organization. We did not vet this video. How could we possibly, every time a video was uploaded? When I used to work at CNN, before you put any social media video on the air, you would do the reporting. Just because it came from YouTube didn't mean that your reporting was any different than the person you were interviewing in an office.

There's no way, nor would we want to do that. We wouldn't want to ask people questions—"Before you upload this, is that really you and your dad going to the beach today?"

It's just not possible. So that's why it's great to see news organizations having people within their structures and their news desks who are looking, vetting.

There's a group called Storyful that's doing a great job gathering material from news scenes, and looking at what's coming in. And then going the extra step, and saying, "Okay, but is this real?" That's just not something that we are equipped to do, or even want to be doing.

JULIA KENNEDY: These have been a raft of really interesting cases. I appreciate all of your participation.

Before I let you go, I want to pose a question that any of our speakers can weigh in on, which is: We've been talking a lot about privacy and free expression and Internet communication technologies today. I'm curious how the types of community guidelines and internal processes that Yahoo! and YouTube have put in place can be applied to other types of businesses and other industries when approaching these types of human rights dilemmas.

ABBI TATTON: I'd rather have advice from you guys. [Laughter]

EBELE OKOBI-HARRIS: This is something that we need too. I was talking earlier about the IC [information communications] techs. I mean if you're in oil and gas, you're used to seeing—I think CSR just has a much older tradition there. So I think within the ICT [information and communications technology] sector this is just so new.

For us this is why being a part of GNI was really important, continually engaging with NGOs is really important—because I think the culture within a lot of ICT companies is very young.

But it's also a culture where people feel like, just by doing our jobs, we're changing the world. But you could think of things, if you did things a little bit differently, what you do actually has a greater impact on human rights.

So I think I would echo Abbi by saying I very much enjoy the opportunity to engage, and to hear from others.

PARTICIPANT: I think there are two areas where this has significant impact for businesses.

One is the guidelines that are introduced in many companies for the whistle-blower provisions, to protect the whistle-blowers, to protect their anonymity, to ensure that they are not penalized for informing the malpractices that are taking place. I think that has had very good impact.

It was very interesting that in the Dodd-Frank bill, that was passed recently, the whistle-blower provisions were strengthened. Perhaps that's an indication that the media awareness—that your organizations have made possible—have now had some business impact in which we are saying, "Let's protect these people, because they are serving a public interest."

Another area is the Foreign Corrupt Practices Act, where the guidelines that companies have in force are very detailed, to ensure that the provisions of those laws are not violated. Again, I think this is an extension of what we are trying to do.

PARTICIPANT: One, I think there's a tremendous amount to learn from ICT, in terms of pushing the boundaries of transparency. So many, many corporations are being asked to be more transparent about their activities, and have not caught up with the fact that people, in general, are sharing much more information now. They're doing it at a rapid pace, and there is a sort of public citizenry of media— which I think is really interesting—in some of the information that you're presenting.

One case that we've worked at, at BSR [Business for Social Responsibility]—I just want to bring up as an example from another industry which might be of interest—working with a very large company that's producing low-cost health-care equipment. Which on its face is a very good thing, bringing ultrasound equipment that's portable and low-cost, into communities, globally, that don't have access to health care.

There was an unintended consequence of that, which was that the ultrasound equipment was being used for sex selection, in India and China, in particular, in feticide for girl fetuses. It was an interesting conundrum, because do you pull the ultrasound equipment from this community, therefore denying access to critical health care? Or do you grapple with the fact that it has this unintended use, and that you—as the company that produces it—are responsible for that?

I bring that up because I think in any of the industries that we work in, there is a real need to think about the fact that I'm producing a product or a service. And I'm producing it with this goal in mind. And I'm focused, and I'm an entrepreneur. Or I'm an engineer, and I'm looking at it from this perspective, but opening up the dialogue to think about what the potential unintended uses of that might be.

We can never know, but always just try to have that dialogue with groups outside of your own product group.

EBELE OKOBI-HARRIS: I think your point about unintended consequences is such an important one. I just want to tell a very quick story also using Flickr.

Flickr, as I said, is used by a lot of people. It was used by a man in Iran who started off, as he says—if you look at his page in Flickr, it says "Flickr made me the person I am today." Because he started off as an engineer, he never was a photographer. He started using Flickr to take pictures. People gave him comments.

He went from taking pictures of his kids—it's the classic, for me, problem of the accidental activist— and he started taking wildlife pictures in Iran. And then the Green Revolution happened. And, boom, all of a sudden he's a journalist.

So he's taking pictures of everything going on. As you heard before, news reporters were kicked out. So his pictures start getting used by all these news organization.

Now, he has a Flickr account that has his name on it, that has pictures of him, that has pictures of his children. And there he is. He has now become a very public person.

When his pictures were first used, he was very excited about it. He posted it on his Flickr page. Although he was upset that Der Spiegel didn't give him credit, that they had stolen his photograph—so there was that issue.

And, so, we connected with him, and we featured him at our Business and Human Rights Summit. A year later he was arrested because of his pictures that were posted on Flickr.

So when you talk about unintended consequences, there is that aspect—okay, this is amazing, he was able to use Flickr not only as a tool to become a better photographer. But he was also able to use it as a platform to create social change. But then, because of Flickr, he was arrested.

So then as a company—I just point that out that we would never have thought of all these consequences. You just don't think—when you create a product, you don't necessarily play out everything.

To your point, it has been incredibly important for us to continue to engage with NGOs and with other people. Because, again, the extent to which and how your products are used, in ways that could be harmful, you don't often know.

JULIA KENNEDY: I do want to give the final comments to Rachel and Susan, who started us off today.

RACHEL DAVIS: My comments, very briefly, would be that I think that there are some real similarities with the challenges we see in other sectors, in other companies. But that, with a lot of what you've been talking about, there are lessons on both sides.

So, policies are a common theme. But policies—and we take it as a given that they are informed by company values—are never going to give you the answer in every case. So the question is always: What processes are there in place, and how good are those processes at responding rapidly?

So back to the escalation question. If you're a mine, you know that if there's a community problem.They block your road and your operations are stopped. That's an immediate direct impact on the company, so you pay attention. For other companies it's not so direct. So that's something where you see lessons being learned.

So what's important is having processes in place that also are capable to this point of assessing new risks—so they're ongoing processes—and constantly scanning what are the human rights implications of our activities, and how do we make sure we're keeping up with them, so that you can then explain the processes that you have in place. If you're not immediately on top of an issue, you can say, "This is how we are addressing it. And these are the types of considerations we take into account."

I think as far as what other industries can learn from you guys, the stakeholder engagement and the dialogue piece is really, really critical. You won't understand the human rights implications of what you are doing unless you are talking to the human rights activists, as you are discovering. So I think that's key.

SUSAN MORGAN: My reflections on this industry, is that there are lots of similarities around the intersection between business and human rights, in this industry and others.

But there are some really big differences too. I think they are probably kind of scale, just the sheer amount of photos and videos.

Also, speed, the speed at which things change. Now, who could have imagined at the beginning of the year, that you would find that issue with the photos in Egypt? And just the speed of innovation, I think, and complexity of the issues.

It strikes me that the important thing is to find a way of dealing with uncertainty, and find a way of institutionally being able to react to circumstances, processes, and things that come up. Whether that's risk assessment, whether it's having a team who are dedicated, whether it's making sure that other stakeholders are involved, so that you've got those connections and those relationships—I think it's finding ways of accepting and dealing with the fact you can't work out every imaginable consequence before you start.

JULIA KENNEDY: Great. Thank you so much for those insights.