I'm Natalie Cone, your OpenAI Forum Community Manager. I'd like to begin our talks by reminding us all of OpenAI's mission, which is to ensure that Artificial General Intelligence, AGI, by which we mean highly autonomous systems that outperform humans at most economically valuable work, benefits all of humanity.

In our discussion today, we'll introduce to the community several important nonprofit organizations currently leveraging AI to influence social impact. By no means is this an exhaustive representation of all the amazing organizations performing this work, but the ones here tonight are forum members, members of our community that have been here with us from the beginning, so we thought it was a great place to start. We encourage you to learn more about the speakers by observing their bios on the event page and exploring the links leading to the organizational pages to which they belong.

The event will be facilitated by Nathan Chappell, founder of Fundraising AI, Chief AI Officer at DonorSearch AI, and the author of The Generosity Crisis, the case for radical connection to solve humanity's greatest challenges. Contributing to the conversation will be Dupe Ajayi of the Ajayi Effect, Jodi Britton, Team for Tech, Allison Fine of Every Org, Anne Murphy from Empowered Fundraiser, Gayle Roberts of Larkin Street Youth, Scott Rosenkrans from DonorSearch, and finally, Woodrow Rosenbaum from GivingTuesday's Data Commons.

Without further ado, please welcome Nathan and the other nonprofit leaders. The stage is yours.

Well, thank you so much, Natalie, and the OpenAI for Nonprofit team. It is such an honor and privilege to be here and to be among so many great panelists and thought leaders, industry leaders tonight. We thought we'd start with just providing each panelist an opportunity to share about two minutes around their experience and or fascination, orientation around AI, and any primary ways that they're using AI today. We'll go alphabetically just to help it go constructively, and then we're going to dive in. We've prescribed a big question for each panelist that we're going to guide through since we have so many panelists. We're going to keep everybody on a really tight schedule.

Let's start with Dupe, if you don't mind, just around two minutes about your experience with AI. Where are you on your AI journey and anything that you think would be helpful to frame the conversation?

Absolutely. I love this question. I have been using AI tools for quite some time now. They are a part of my everyday life. Everything from timing myself, the Pomodoro method, if you guys don't know, get on it and in it, all the way to script writing, thinking about ideas, brainstorming, transcribing meeting notes, all of the things. These tools help me to be able to move faster, quicker, get things done in a way that just feels more natural to me. Of course, then I add my own sauce to it. What I love the most is that over time, these machines, these platforms have gotten to know more about my voice. They're answering in a way that is unique to me. It's really been a delight and I use them almost every day.

That's amazing. Yeah, I love that. Thanks for leading us off, Dupe.

Dr. Jodi Britton, would you share? I don't know. You've earned it. I feel like I need to call you doctor. Okay. Yeah. Thanks for doing this, Natalie and Nathan. Thanks for having us. Team Protect is a non-profit impact accelerator. We work specifically with education-focused organizations globally, most of which are working in very under-resourced communities. Right now, we have about 800 non-profit organizations from 90 countries in our community. We use AI all the time. For me, I think the thing that is not only our responsibility to get informed about, but our responsibility to get excited about, is how AI can really be used as a lever for change with all this. As we talk with our non-profit partners, they want AI to be part of their story. The work that we're doing right now, mostly with AI literacy, is creating a foundation with that story that's going to unfold. It's a pretty exciting time around our work right now. We are doing a lot of work through a Gates-funded initiative, AI4Education.org. Through that, we're actually creating AI tools, things like offline LLMs, multimodal teacher coaching, things of that nature. We're also internally doing a lot of streamlining of content curation, working on all of our megaprompts. We're using AI to analyze really large datasets. We're getting a lot of actionable insights that can really help us accelerate impact with more intention, I guess, if that makes sense.

That's amazing. I feel like coffee in the future, to talk about megaprompts, is something that has to happen. I'm definitely going to be beating down your door on that.

Let's move to Alison. Alison Fein, do you want to share a little bit about you and your work in AI?

Sure. Thanks for having me, Nathan and Natalie. It's a pleasure to be here. My career has been spent at the intersection of social good and digital tech. I've written four books on the topic. The last one was called The Smart Nonprofit About the Use of AI for Social Good. To me, the promise of AI is to undo the busyness of our last chapter in digital tech, where we got bigger and faster and much, much louder, and where most of our staff people and nonprofits spend most of their days in front of their screens, which isn't good for them and isn't good for the causes that they work for. I joined Every.org 18 months ago, because we have an opportunity to remake the field of fundraising by helping donors to find causes that suit their interests better, to make that discovery more interesting, more joyful, and most importantly, to make the intersection between donors and causes more meaningful as well, because back to the last chapter in busyness, the field of fundraising has adopted a very transactional approach to fundraising, where a donor is much less likely to give a second gift than to just go away. Much, much less likely. Because of that, it creates an internal panic, and everyone's on the hamster wheel of sending out all the emails we get, all the mailings we get, and the burnout rate for development staff is sky high, and those donor retention rates, as I mentioned, are very low. And it's really causing just so much harm to the sector. So I think AI has enormous promise to create a new chapter in fundraising that is more joyful, that helps to sustain all the causes that we care so deeply about, and lets us to be actually more, rather than less human, in this next role.

Yeah, that's fantastic, and I can't wait for you to dive into your work and your question. You have some really practical advice for people. Let's move to Anne Murphy, who is, I think, a recovery and fundraiser that had a lot of love affair with AI and was just loud and proud about it, so. I am, I've kind of become the Tony Robbins of generative AI in fundraising for better or for worse. I think, you know, as a 25 year veteran of frontline fundraising, there were two things I was always obsessed with. One was, there's got to be a better way to do this. Like everything we do, there's gotta be a better way. There's gotta be a better way. Just like a lot of other Gen Xers, of course. And then also just the leveling of the playing field. So a couple of people have already mentioned that, but how generative AI can level the playing field for all of us in the nonprofit sector. And I just keep getting all this great reinforcement. So I put myself out there as the fifth grader to the fourth graders to teach people how to use AI. And, you know, my programs are populated almost exclusively by women. Like we're all here to level the playing field and it's really exciting and I cannot get enough of it. And I've never given up so much sleep so happily ever before. Thank you for having me. Thank you OpenAI and thanks Nathan. Yeah, I love it.

All right, well, Gayle Roberts, we met each other a few years ago. When I think about you and AI, I think about the Nike logo of just like AI, just do it. Because you're one of the first people that jumped in the deep end of the pool and just went for it. And so, yeah, share a little bit about you and your AI journey.

Well, I jumped into the pool probably because I'm a swimmer and I feel best in the water. I have a mermaid painted on the side of my house. So I'm a Chief Development Officer at Larkin Street Youth Services here in San Francisco where our mission is to end youth homelessness. I've been with the organization for the last five years and under my watch, my team of mighty fundraisers has raised $75 million in private funds. And increasingly that is being leveraged by AI. We use Salesforce as our basic CRM and about three years ago, we started to make it more AI enabled. I use iWay, which is a research tool for predictive for wealth screening. And then we also use Gravity, which is a major donor pipeline management tool. Pings me every morning, tells me who in the database is the best prospect, writes me the email, does that for me and all my solicitors. Other tools that we use that are becoming more AI enabled include HubSpot and Asna and others. So we definitely have a applied AI stack. Of course, we also use Generative AI where I'll use OpenAI's chat GPT where the long time subscribers and then use it for all of our content creation, including grant writing, project planning, marketing content creation. I use it in the morning for my personal journal. I mean, like I use it for everything. I'm like Anne there. And then on the programmatic side, we've been in part of a controlled research, research controlled trial for the last three years, three year project. We've been in it for one year to demonstrate the effectiveness of basic income. This is supported by Google. It's being run in six cities across the country. Chicago University is doing all the data review on it. And we're pretty confident that we're going to show the effectiveness of basic income for young people who are experiencing homelessness and possibly for others. And then finally, we're in the midst of also developing an AI enabled co-pilot for our case managers. I've got some money from Tipping Point Foundation. I'm trying to work with Amazon on that potentially as well. And that will be an assistant for all of our case managers so that potentially looking at real time and historical data, it can find the youth who need intervention maybe before they even notice it and then help them with the project plan. So we're all in. Yeah, no doubt. Yeah, it's amazing. I mean, I'm pretty much fully immersed. So that's awesome.

Scott, let's go to you and share a little bit about your work in AI.

Wonderful, thank you. Thanks, Nathan. Thanks, Natalie, for having us. I'm Scott Rosenkranz. I'm Associate Vice President at DonorSearch AI. We build custom machine learning models for nonprofits of all shapes and sizes to predict who's likely to make a gift in the next 12 months. So we say that we quantify the dynamic connection that every individual has with that organization, knowing that most nonprofits, like Allison mentioned, have way too many people, but not enough resources to effectively manage all those relationships. By telling them and prioritizing who they should be getting in front of, who wants to engage further and stronger and have just a more supportive, engaging relationship, and also telling them who not to, who do they not have to prioritize and waste money, waste time, waste limited resources by chasing the wrong people for the wrong reasons. So that's what I do in my day-to-day. I work with a wonderful team of data scientists, but I'm also a member of Fundraising AI with Nathan and some other members of this panel, where we talk about, and we've created a framework for responsible and beneficial fundraising for responsible and beneficial AI for fundraising, knowing that responsible is necessary, but so is beneficial. Nonprofit sector works differently than all the other sectors. So trust is a vital component, and we need to make sure that it's not just having AI be efficient, but it has to be valuable, beneficial, be doing the right things, and again, promoting that trust. And then I co-host the Fundraising AI podcast as well. So I'm an early adopter of AI. I use it all the time to my wife's detriment. She hates her smart home, and things don't work the way they're supposed to, but I'm, again, excited to be here, and a huge early adopter of open AI, chat GPT as well.

Yep, I know you are. We talk about it every week or pretty much every day.

All right, so last but not least, Woodrow Rosenbaum. We've known each other for a couple of years, and you just are an amazing voice and advocate for generosity. So share a little bit about yourself.

Thanks, Nathan. Thanks, Natalie. It's good to be here. It's a great panel. So I'm Woodrow. I'm the Chief Data Officer for GivingTuesday. I manage our data engineering and data science teams in a global collaborative effort to build research and data products for the social sector broadly, to understand the full spectrum of human generosity and its impact, and to help nonprofits to navigate that and make data-driven decisions and improve their outcomes. Last year, we had partners asking us to help leverage our data assets to support the development of AI tools and to convene in the GivingTuesday Data Commons work around developing AI solutions for the sector, and we launched the GivingTuesday.ai, which is our generosity AI working group. We have a little over 900 researchers, developers, companies, practitioners in that working group finding solutions to problems that are identified by the nonprofit community around the world to ensure that we are building those solutions by and for them. Thanks for having me.

Awesome. Well, thank you all. And I'm tracking time. We've got a big panel, and we're gonna try to get through everything, and we're doing really well. You all get an A, gold star for every person here for helping keeping us accountable on time. I wanted to provide, as we get right-

before we tee up the individual questions and create some of that narrative, I want to make sure that we provide a backdrop and such a just essentially a lay of the land of where we're at in the nonprofit sector. Because I know anytime we do a session like there's there are people that come from all different vantage points. Maybe people have been nonprofit sector for one day, and maybe they've spent their entire career. So I think it's good just to level set.

So I'll just provide a couple statistics and share just a few of nuances around where philanthropy and the nonprofit sector is today. So for those that don't know, philanthropy in the philanthropy sector, which are a lot of times, there are many different types of nonprofits, but those specifically that are classified as a 501 c three organization, it's about 1.7 million of them in the United States and around 10 million globally that kind of meet that same, you know, nonprofit NGO kind of definition.

These are organizations that are exist to essentially fill gaps that either corporations or governments can't do on their own, and oftentimes, are essentially there to create in or to reduce inequities.

And so, you know, when I was hardworking people that wake up every day with this notion of making the world a better place, it's the second largest source of US employment. Overall, not just in fundraising, but in overall, if 5.6% of the GDP 1.4 in nonprofits contribute 1.4 trillion to the economy in 2022. For those that are on the fundraising side, not the programmatic side, but the fundraising side, fundraising represents, it's just roughly a little bit over half a billion dollars. I'm not half a billion, half a trillion dollars, I missed a zero there, half a trillion dollars in in in the industry.

So it's an area that is robust, it's extremely nuanced, it's very difficult. But it also is an amazingly rewarding area to be in. It's also facing some really significant headwinds. The sector is changing and has continued to change. For the last bit, we've seen a dramatic increase for services with, you know, record inflation.

So nonprofits are essentially taxed, not in the physical way, but taxed in the way of like, required to do more, asked to do more to serve more. Over the past several years, we see a significant increase in burnout from a staffing perspective. So a lot of latest reports, in fact, I was reviewing one earlier today, the number one concern for nonprofit leaders right now is staff burnout and workload, overload.

And then we've also seen, you know, systemic decreases in the percentage of households that give to charities. So Allison's written about this, I've written about it in the generosity crisis. But just to, you know, share a little insight on that, is that 20 years ago, when you were walking down the street, it would be extremely practical that two out of three people you'd meet on the street were giving to a 501c3 organization.

But fast forward just 20 years, that number is less than half. And so dramatic decrease in the percentage of people that are expressing their generosity through a 501c3 organization, they might be expressing their generosity through other vehicles or purchasing responsibly or through GoFundMe. But just that the 501c3 organization, the money that's going to them to help them serve those missions, that increased demand for their missions, is providing a lot of strain.

Now all that to say is that the need for innovation in the sector has never been more important. That we can sit here as a word I often use now is a hopeful pessimist, because like we know these concerns, and we feel the effect of those concerns. But we're also hopeful because we're living in a time where a transformative technology has arrived that can absolutely change that trajectory. That in fact, AI in its forms, we'll dive into different forms later, in its forms is really the only scalable solution to reversing that decline. And so we see this tremendous amount of opportunity and enthusiasm for AI to give nonprofit wings. We'll just keep on stealing from Nike and now Red Bull in this idea that AI gives nonprofit wings.

We also recognize that AI adoption is in the nonprofit sector is remarkably lower than the for profit sector. And there are lots of reasons for that that we won't get into all of today, but perhaps another session that we can do to talk about some of those barriers, but that we have somewhere around 12% adoption of nonprofit of AI within the nonprofit sector. We absolutely have a lot of work to do.

And I think in platforms like this can help encourage people and we'll dive, we'll peel back this onion. No matter where you're starting, there's somebody here on this panel that can help you think about things differently.

We also have to do this, as Scott mentioned, and I know many others have already talked about responsibility, is that we have to do this differently. The nonprofit sector is in the business of trust. And therefore, the framework that we evaluate AI for nonprofit organizations has to look a little bit different, that first and foremost, it has to prioritize trust in everything that it does.

Because if we diminish trust, you know, it's one of those things that's very hard won and very easily lost. And so we have a higher role and responsibility to use AI in a way that is going to increase trust within or between individuals and organizations. And essentially, this is a great opportunity for our sector to lead the world in what that actually means.

So today's panel is we're going to dive down, I've asked each panelist to really think through kind of a big question that relates to their subject matter expertise. When I, when we've met, and we've did some interviews to really get to this point, it really came down to a couple key themes. And those key themes were really on the use of AI around personal time. And you're going to hear from Allison about that. And also from Anne, we're going to hear about programs and operations from Jodi and Dupe. And as well Scott, fundraising with Scott and communications is where an area that Dupe is very deep in that communication side.

So we're going to cover a really broad swath today, knowing that this is kind of the first of what hopefully will be many other deeper dives into some of these topics. So we're going to cover a lot of ground in a really short period of time.

And with that said, if everyone's buckled in and ready to go, we're going to start at the macro view. So really, I don't think there's anyone better to talk about the macro view than Woodrow Rosenbaum, because he thinks a few years ahead of most people. And you're like, Wait, Woodrow, where are you going? Because he's already planning on things that people maybe aren't ready for, but they're going to be ready for.

And so thinking through kind of that macro view, I mean, Woodrow, we've known each other for a while. And I, you know, can't think of anyone better than to tee this up, is that when you're thinking about this very broadly, AI nonprofit sector, what do you think the nonprofit sector needs to realize the opportunities and potential of AI?

That was very kind, Nathan, thank you. I think about this in a couple of ways. First, I think that there is the obvious and continuing challenge that we have in the nonprofit sector of keeping up with technology. And we've seen time and time again, this sector get left behind. And that's, that's the sort of expected situation, I think that a lot of people are, are facing this, "oh, are we just going to be behind the capacity curve yet again?"

The flip side is I think there is more of an embracing of the opportunity of this new technology now than we have seen in the past. There is a lot less kind of inertia in the system. And I think that that's an opportunity for us to, as you said, to lead the development of these tools. And that's the big, big difference here. And if it's less in fact about missing the opportunity of adopting the tech and more to my mind about the potential opportunity of being able to drive the development of this, the needs in the social sector are quite challenging. And the potential outcome of serving those needs is one of dramatic, positive social impact on many, many domains.

And if we embrace the technologists that are developing these tools to solve these problems first, not as an off-label solution, but as the primary purpose of these tools, then the tech itself will be robust. And I think we need three things in order for that to happen. First, we need a place for some transparent experimentation. We need to be able to test how these things are working and see, really interrogate how these solutions are solving validated issues for the social sector around the world. Yes, including in fundraising, but also in just how organizations meet their missions.

Second, we need assets to support that development that are specific to the nonprofit sector. We need to make sure that we have the basic building blocks of these that on which AI solutions are built that are sourced, ethically sourced by and for nonprofits, specifically for those use cases. And that those solutions that we are then interrogating in that experimentation environment are built with those assets at the core.

And then the third thing we need is a new framework for capacity building. Implementation of these tools against those validated use cases. Again, where we have an enormous opportunity here because if we build something that serves one purpose, it becomes very extensible if we're doing that collaboratively. And in doing that, we really need to be very clear that we avoid the risk of just amplifying existing bad practice. And I think if we can keep our eye on that, if we can work collaboratively, and if we build on top of assets and a tech stack that is purpose built for the social sector, then we can meet that opportunity of guiding the development of these products rather than keeping up with them.

All right, that's amazing. I mean, you've covered so much ground in three minutes, Woodrow. Thank you so much. And a huge shoutout to really GivingTuesday as technically a nonprofit that is creating a series of resources that are open source for people to grab data, to explore use cases. So really wanna encourage everyone to really check that out and lean in and join, be part of that community.

So let's move over to Jody. Jody, you have, even in your intro, extremely deep and experiential use of AI in kind of a wide variety of things. And so it's where you've been focused on, you spent a lot of time on AI literacy, capacity building, you talk about workflows, I mean, kind of immersed in so many different ways. One of the things that you've shared before is that you see AI being a force for equalization. So I was hoping that you could share a little bit more about what that means and how that, how nonprofits can use AI as a force for equalization to support their mission in a really powerful way.

Yeah, absolutely. Woodrow, I just loved everything that you said and it resonated with me so much. I think in our work with nonprofits, we've really observed that AI kind of holds that potential to either be the pretty significant source of inequity or a powerful equalizer. So our aim as an impact accelerator is to really ensure that this achieves the latter. Like we wanna make sure that even the smallest organization can overcome resource limitations and enhance their mission delivery. All of this is about mission for us.

I think traditionally small nonprofits have struggled with things just from lack of resources. It can be anything from copy editing to accounting to grant writing. We're experiencing for ourselves and with our partners, like how AI tools can really give us more time to focus on mission critical work through our AI for education initiative. Like I said before, we're developing solutions but we're not just developing solutions to develop them. We're building with and for the communities that we serve. So we're looking at very specific use cases, very specific solutions and trying to figure out how not only we can build those things as one-off deployments, but also build those things for sustainability across organizations.

I think just in terms of like local implementation and sustainability as an accelerator program where you really wanna see overall impact with educational outcomes, because that's our focus area. But we also wanna see growth and sustainability in scale of those organizations. So one of the things that we've kind of seen happen before.

One second, can everybody that's not Jodi, please mute? There's a microphone icon in the bottom center of your panel. Thank you so much guys. I'm sorry to interrupt you, Jodi.

No, you're fine, you're fine. I thought maybe one of my kids was just talking to me. Natalie, it's okay.

No, I think we're trying to think about things like just staff turnover, right? So how can we create closed libraries of training materials and chat bots so that we can free up that one-to-one human interaction time to be really meaningful and really focused on the pedagogical work and free up those kind of small struggles in ways that we couldn't do two years ago. Now we can. We're also really thinking very carefully about things like personalization of program materials. 90 countries in our community of practice, we have to figure out what's gonna be the silver bullet through kind of the core mission to hit a lot of different organizations. So just thinking about how we can use AI, where they used to pay consultants or staff exorbitant amount and wait years to be able to just adapt programs. Now they're able to really make some cool things happen just with AI, whether it's brainstorming, whether it's reimagining, whether it's refining, all those things are happening.

A lot of them are just using Chat GPT and the GPTs that are kind of publicly created in that space to get that work done. Next week in collaboration with New America, our AI hackathon there is really gonna be focused on open educational resources and ensuring that our nonprofit partners, when we bring that back, we'll be able to overlay those AI tools so that they can find, adapt, reuse and access programming materials and resources immediately in some ways that we have not been able to achieve before. We've never been able.

to get there. So, you know, we're really focused on recognizing the challenges that are faced by nonprofits and they're not new, you know, but what can we do as an impact accelerator, you know, to really push things forward. And we're seeing firsthand right now how crucial it is for the nonprofit community, not only to develop that AI literacy, but also to share their tools, you know, quality assurance measures, data sets, all of those things.

And it's a pretty amazing thing to witness, right? Because combined, it's like accelerating the life cycle of our impact. And it's going, you know, it's going pretty good so far. We've trained over 300 people, 300 organizations, and just using chat GPT effectively and creating tools every day to make sure that they can do more with it.

So, you know, Nathan, going back to your question, I think from our perspective, you know, we're really trying to leverage AI to kind of ultimately enhance the work that brings each of the unique missions of every organization we serve to life. That is amazing.

And I hope people in this forum connect with you on, you know, this idea of just jumping in and trying these things. And it's what amazing to me to think about right now, everything that's already been shared, and we have much more to share, is like all happened in a relatively, I mean, actually a realistically very short time. We're talking about transformative shifts in a very short time. And I love that. I think there's so much power in learning from each other. So we're going to continue to peel back the onion a little bit more.

Dupe, you have a lot of experience in, you know, marketing, communications, both in the for-profit sector, the nonprofit sector, and kind of everything in between. And, you know, as we think about one of the biggest shifts for society and business in general, is that we are inundated by information, like just so much information all the time. And what used to work no longer works. So people are distracted, they have shorter attention spans, the competition for connection is real. And throwing spaghetti on the wall, you know, just to see what sticks is no longer a viable or efficient strategy.

And so, you know, what I really like when I've been researching your work, we haven't known each other very long, but I like I feel like we're going to be friends for a really long time, is that your expertise and passion is really around the area of using marketing strategies to be inclusive, to reach diverse and underrepresented communities.

So I was really hoping that you could share how AI really can be leveraged to do that, to really, you know, market to help nonprofits, you know, have a voice, be able to communicate effectively and through the noise and, and advocate, but not just broadly, but within those underrepresented communities.

Absolutely. And I love this question. I could talk about this all day long, but we will get it done in a few minutes. You know, when I think about the possibilities, when it comes to say, for example, disaster response, and thinking about AI, assisting in the decision-making process that has to happen very quickly, when people are in an emergency, right, this is a very real-world application, it's a very, you know, pure example of how you can use these tools to find out, okay, where is there a safe space? And where is there, you know, for example, where can people be housed, and that type of thing, you know, it's a very obvious solution.

But I'm thinking about, you know, the collection of stories and storytelling, right? So behind all of the infrastructure of AI, and the work that our incredible organizations do, there are the stories, and that is what pulls people in, that is what drives our fundraising, that is what gets people to support, that's what gets people interested. So imagine this, imagine if you have, for example, hundreds of stories, for an international nonprofit organization, and you have been collecting them over time.

This is a pure example of how you can use AI to sort through, look through those stories that are going to resonate most for a campaign. So say, for example, we're talking about a campaign connected to cheddar cheese, it's late in New York, and I'm hungry. Here's an example of where you can look through very quickly, all everything that's in your story library, and surface those stories that are related to cheddar cheese, and knowing that they are going to relate to people who enjoy cheddar cheese, funders, etc, and the like.

So for me, that's a very, you know, pure example. It's a very easy example that any organization can employ today. And then we look at voice, and I cannot over-stress this, I cannot over-emphasize this, we were talking about this as we come on board, there's a special type of delight in hearing a voice that sounds like you, seeing someone that looks like you when it comes to something that you're engaging in.

And that goes, it's the same case for our constituents that participate, that we serve through our nonprofit organizations. Here's an opportunity for us to really lean into the texture of individuality when it comes to engaging through marketing. I, earlier this week, saw an article in Tech Crunch that was mentioning some Black founders, just nuancing chat GPTs for tone, specific to African American voice. And that's exciting, right? Because here we are, where we can drill down, drill in, and pull through intersectionality in a very pure way, right?

This is something that should be done. I think it's the standard. What was interesting was that in the comments, and there weren't many in this Tech Crunch article that had an issue with this still. And so as we're thinking about, and they were wondering about this being useful, history has shown us time and time again, that when you lean into the individuality of people and who they are as humans, you benefit, right? Across the board. And here is a way that through marketing, you can very easily drill into who it is we're trying to speak to, and surface that, and pull through in all of your campaigns. It is quite evident. I'm excited because we're not having to look through hundreds of stories of individuals and having to do that work. And we did that work very well at City Year. Now we have the advantage. So there's that. And that is my time, Nathan.

Well, now you've got me hungry. I'm going to order a cheese pizza after this. But you're referencing something that would have been very difficult, if not impossible in the past, right? Very human, labor-intensive. And you're saying now it's at your disposal. Now you have this ability to do this thing, not effortlessly, but much in a greater capacity, which I think is just truly amazing. Absolutely.

Yeah, that's fantastic. All right. Well, Scott, you and I have known each other for a while. We've been working in AI since 2017 together. Your work, obviously, I mean, you have your degree in psychology, really approach machine learning and deep learning from the essence of what are people thinking? What are their motivations? So, you know, but knowing that people are coming into this call, probably largely thinking about AI.

AI as being a ubiquitous term with generative AI, right? And your experience is a bit more on the predictive AI. So I thought it would be good for this panel because knowing that people come in, they may not know anything about predictive AI or what the differences are. Do they compliment each other? Do they contradict each other? So I was just hoping you could give us that three minute primer on the differences between predictive and generative. And then also kind of this idea of this term that we've used around precision philanthropy, which is an area that I think we should be really optimistic about.

Yeah, happy to do so. Also started my timer to make sure I'm good on time. So predictive and generative are two different sides of the coin, but they work incredibly well together, right? The reason that AI is so big in the space, thanks to open AI in November, 2022, is because it works differently than predictive. Predictive is generally indirect. It's what Amazon uses to recommend the things that you should purchase, what Netflix uses to recommend the shows you should watch, what Google recommends, uses to recommend what route you should take to get through the city. You don't have direct control over it. It's really good at objective, at logical, at quantifiable information, a yes or a no, do this or do that. This person will respond this way versus this way. Whereas generative is more open-ended. It's more creative. It's more subjective. And also, again, as a result of Chats with the T, it's more direct. I have direct control over getting the output that I wanna get by just changing my prompts. I don't have direct control over Amazon's algorithm and predictions of me unless I'm very specific in what I purchase to control that model, right? So there's two different ways to interact with them, but when you put the two of them together, it's like one plus one equals five, right?

A lot of people focus on the capability of generative AI to, especially in the nonprofit fundraising space, to reach out to more people quicker, right? But predictive AI allows you to reach out to better people better, right? So going back to Dupe's case, when if there's a campaign for cheddar cheese, we don't wanna send it to people who are lactose intolerant or vegan, right? Because at best, you won't get a gift. At worst, you're gonna piss off a lot of people, but you wanna send it to cheeseheads, people who take tours of cheddar cheese factories, right? You wanna make sure you're focusing on the right people, not to make this open AI's forum meeting on cheese, but you wanna make sure that you're talking to the right people and you're meeting them where they're at and where they're interested, especially nonprofits do not have resources, unlimited resources to reach out to everyone. So that precision philanthropy, we pulled it from precision medicine, right? Treating each individual and their disease as an N of one, knowing that cancer is not just global cancer and affects everyone the same way. Philanthropy is not the same for everyone. Engagement is not the same for everyone. And so we wanna make sure that we're seeing what the motivation is for each person to continue to engage in a certain way. How can we intercede and receive them where they are and strengthen that relationship while at the same time, again, knowing who to put aside for now, right? There's no shortage of information delivery. We're getting way too many emails, way too many messages, way too much content online. So it's not like we just need to push out more of that to cast a wider net. We just need to cast a better net. And so that's where we've really prioritized that predictive component, but then using the generative to help and assist with that outreach. So that way you can offload that and utilize that resource as well.

That's awesome. And I have a feeling that we're gonna be talking about cheese on our next podcast. So thanks for that. Scott's good at coming up with lots of analogies around things that are very abstract. All right, so it's time to get pretty practical. We're doing great on time. We've got a great flow. We're peeling back the onion, going from macro down to micro. So Anne Murphy, I already gave my thoughts about your early introduction of this accidental or either recovering fundraiser or fell in love with AI. When I think about you, if Gail is the AI, just do it person, you're the AI, I'm loving it person. This exudes everything about you. And I think what you've helped a lot of people overcome is that initial fear of like, where do I start? Like, what do I do? It's just like, it's this black screen and this little prompt and I don't understand it. What do I do? And you have tremendous patience and I think passion for training people. So let's just dive in. So my really practical advice, if there's people on this, hopefully people know how to log into chat GPT that are on here. So maybe not tell them what the URL is because let's assume they're all here. But how do individual staff get started with generative AI and really around, I want you to share a little bit about that in some of the tricks that you've come up with in your trainings, but also your thoughts on making sure, 'cause I know something you're really passionate about, making sure that's done in a responsible and beneficial way.

Yeah. Oh boy, I love this subject so much because there are so many folks sitting on the sideline because they're afraid and because it's new and because they don't want to make a mistake and they don't want to technology themselves out of a job. And I know lots of those people and I have been one of those people. So I'm gonna share like our three steps that we do with people as they're coming into our programs to get familiar with AI and then eventually move toward adoption and leadership in their organization. So there are three main things. One is a commitment to an abundance mindset. That's really, really important. Mindset is critical. So that's one. Number two is to learn through use cases you actually care about. The people who bring use cases to the table that are on topics that are important to them or very, very important to the world tend to adopt things faster. And then the third is to develop your own personal code of ethics before you move into doing that for your enterprise. So within the mindset piece, there are three things that I always talk about that are essential. One is having a curious mind. So not a technical background, not necessary in some ways, not helpful. This is the revenge of the liberal arts majors. So a curious mind, having a willingness to be vulnerable. So we all high achieving people, we love to do things we're good at. We're asking you to do something you are not good at and nobody's good at it at first. Nobody knows. These things don't come with like rock solid operational manuals on how to do this. And then the third piece is community. This stuff is so like face meltingly confrontational to our self identities that having friends, work besties, other nerds who can sit around talking to you about, oh my God, I can't believe this thing that I discovered that ChatGPT can do. What have I been doing with my whole life up until this very point? So those three things, curious mind, willingness to be vulnerable and community are really important. What I've seen with people talking about working with use cases they love. And first of all, I'm willing to go to my grave on the whole, you learn better through use cases. Use cases is how we get dopamine. We adult learners need our little pellets of dopamine to power through annoying technology gaffes. So I believe use cases are the way in. That's how people get their light bulb moments. We did a 24.

our chat, custom GPT building challenge. And what was so cool is we had people coming from like absolutely zero, zero to 60. But they were bringing use cases, you know, working for the Alzheimer's Associate doing a GPT for the Alzheimer's Association for Susan Coleman race for a cure, things that affected their family. And we had people who had absolutely no business, be knowing how to do this stuff. Within an hour making custom GPTs that made them really happy. So that's important. And I always encourage people if they, if they can't figure out the use cases, just DM me DM anyone on this call. If you're wondering, like, what can I tell a nonprofit colleague to actually do with AI, that's going to be exciting. Just ask any of us, we can we can help you with that. And then regarding a personal code of ethics, which I think is your first step to AI governance is just my first step is put an hour on your calendar next week, pull up the fundraising AI framework for responsible AI and fundraising, and reflect on how the work that you've been doing so far with AI bumps up against some of those things in the framework. And that's the first meeting of your AI Governance Council. It's just you. And then you go from there. Awesome. And you covered a lot of ground in a short time, really, really appreciate it. All right, we've got two more to go. And we're going to move to Alison, we're just getting super practical. You know, it's this idea of just do it, try these things. But Alison, we also know that, that AI transformation, AI adoption is a journey, not a destination. And you've been you're a prolific writer on this. And you talk a lot about the need to keep humans at the center of our field, humans at the center of AI. And really, the success of of AI adoption should be measured in that sense that humans are are in the loop. One of the areas that you've talked about that resonated with me is really where the rubber hits the road. And I it stuck with me the first time you said it. And I repeated it many times since is really around your ideas to move away from the scary but that AI provides as a dividend of time. And so I was hoping you can share this because I think it will stick for a lot of others as well as like, what is the dividend of time? How important is it? And how do you enable it? Thanks, Nathan. And I want to give a shout out to my co author Beth Cantor, as well. And we worked on that idea together. So this is the return on investment. And so many people as this entire session is about are tiptoeing into the AI waters right now. And the question is, why are you doing this? Because everyone else is doing it because the nonprofit down the street is doing it. And what do you expect to get back from that? Right? So imagine you're the ED, you're in front of the board, you're starting to talk about we're going to use AI. And my fear is that you'll do it to supersize existing activities, right to do more of the same. And that's not a great ROI because we know the systems aren't working well right now and people are drowning in administrative tasks. And all of the bad data around current fundraising, Woodrow brought up you brought up at the beginning as well. So there's another way to think about this. AI has the capacity as the most powerful technology regular people have ever had at their fingertips, to greatly reduce the administrative overload of people within organizations, right, the workflow, the finances, the communications, every task that's taking up a task that's taking up 30% of people's time, inputting data, say, that can be greatly reduced, AI is great at doing rote tasks. And then you have this dividend, right, you have this return, which is time. Now imagine what you could do differently, if you could get up from your desk, and look at the world, and say, I want to be in relation to people who care about our cause, I want to pick up the phone every day, or go out to lunch every day with a $25 donor and say, what does our cause mean to you? Why did you come here? What's your story? Right? Well, how can we make this about you, instead of about our busyness about our to do list, which is what so many fundraising appeals are really about all the things we have to do. And imagine how joyful that would be for staff. And imagine how different the experience would be for all of our stakeholders outside of our walls, to now actually know the people who are doing this work, to be able to reach somebody if they have a question or want to share a story, to connect them to one another in your communities, right, to pray by zip code, set up some coffees for people just to get to know one another, right, that the dividend is about how can we be more human in this next chapter of our development? Not how can we suck the humanity from all of our work even more. And I think Nathan, people are dying for human connection right now, people are lonely, people are looking for meaning. If you run the organization that is deeply invested in connecting to people inside and outside of your walls, helping them tell their own stories, you are going to win this next chapter hands down. I love everything about that. I'm sure everyone does, because I think we all feel it. And we all know that, you know, humans are such an integral part of the success of the nonprofit sector. And if I had, you know, I spent 20 years leading the nonprofit, if I had, you know, a Gen AI, an Open AI, Chachi BT at my desk 20 years ago, I mean, imagine what I could have done to channel that extra time into that personal relationship. So much promise and opportunity there. I hope people really hold on to that.

So, Gail, you know, I think, well, they say that we saved the best for last. I've been on several panels with you over the last, since November 30, 2022, when Chachi BT came out, somehow the universe kind of connected us. And, you know, and I've seen you and your AI transformation evolve pretty dramatically, just like, you know, just start this one thing and then another thing and another thing. And now you're seeking new and innovative ways, you know, to R&D into this. And you've also, your thoughts around AI have actually evolved as well. And so, you know, you've already shared, just do it. I think that's probably Gail philosophy, you know, number one, just do it. Don't be afraid, just try something. But really want to, you know, as we land this plane almost back full circle to where Woodrow started this kind of more of a macro, like how can the nonprofit sector, which includes, you know, nonprofits and lots of other types of organizations, work with, you know, to ensure that the futures, the AI today and the future of AI, whether in AGI being, you know, the clear mission of open AI, make sure that it reflects and benefits all of humanity, and also amplifies the voices and the needs of marginalized communities. It's an area that you're really outspoken in and a real strong advocate for. Do we have all hour? You've got four minutes. I could write a book on that, right. But I think Nathan, you already have, in some ways, you know, just to kind of quickly summarize some of the things that we've been hearing, you know, Nathan, you did talk about how the nonprofit and

geosectors are in this country, at least 10% of the economy and globally, a significant part as well, with almost 2 million people in the US working in this sector.

Woodrow talked about the importance of investments and pilot projects and new frameworks for how we do this work.

And Judy talked about how AI can really enhance our missions and effectiveness and reach.

And Dupe talked really about what I got about the personalization of marketing so that everybody can see themselves and have their voices be heard.

Scott, bringing together both the predictive and the generative so that we have more precision in our work, particularly around fundraising and development.

And Anne, I also heard you say, really, how do we help people take those first steps to get more comfortable with this technology?

And Allison, this promise that we'll all have more time on our hands. And I'm like, God, I hope that happens. I really hope that happens. I think in the short term, maybe that's not true.

So I think the main message I have summarizing all of that is the AI sector needs the nonprofit space at the table.

I have heard in the last year and a half very little conversations that include nonprofit leaders in what the future of artificial intelligence, whether it's predictive or generative or AGI or ASI or whatever, where we're going.

We have not only huge communities that we represent, but also extreme expertise and a big back bench that can be leveraged in a variety of different ways.

And as we think about the future and some of the challenges that we're going to face, we need to make sure that communities are involved.

And the more that we can engage our communities in advance to get everybody facing the right direction, I am concerned about some of the possibilities of the future, including just the social disagreements around what this technology should or should not even be used for, let alone some of the more potential risks that are down there.

So I think the AI sector could really benefit from the nonprofit sector to create community partnerships so that these tools can not only be leveraged effectively, but will actually be embraced and used to make the greatest good in the world.

And yeah, I'm a fundraiser. And so yes, there needs to be resource development. There needs to be pilot projects. There needs to be technical assistance. And we're under-resourced, under-capitalized, and fragmented as a marketplace.

But there is the opportunity to make dramatic shifts over the next couple of years in our own lifetime. And I'm excited about what the future might hold.

Always got to make the ask, Gail. Great job. All the fundraisers will be proud of you for doing that.

And fully agree, the nonprofit sector, with the lack of financial incentives that we have to scale things, which the private sector does, we are a great group and a body of group to enact change at scale with AI.

So Natalie, we're at 802. I think we did the almost impossible. We did with this amazing group of talented individuals.

So thank you again so much for providing the space and the platform for us to talk and represent the nonprofit community within OpenAI for nonprofits. So really, really appreciate it.

Thank you so much, Nathan. And everybody here today, it is truly an honor to have you. And as Nathan and I continue to say, this is just the beginning. We're just planting the seeds. And we chose to highlight you all because you've been here with us from the beginning in one way or another.

So now we're actually going to facilitate a 20-minute Q&A with the audience so that they have an opportunity to surface some of their questions.

And we are piloting a new method for curating our Q&A portion of the event. So if you haven't voted yet, please vote on the questions that you'd most likely hear addressed.

And as of right now, our highest voted question comes from, I know it's down here. Oh, okay. Kavita, Kavita, would you like to unmute yourself and ask the question to the panelists?

Sorry, I didn't expect that. But I think Dr. Jodi, you addressed some of this question already in your conversation.

But especially I'm curious about in the nonprofit sector, how are you re-skilling your existing workforce, right? Because they are the most closest to the problems. And like Annie said, they're the most willing and able to bring the use cases to solve with AI.

But I would be curious, how are you organizing? How are you bringing the use cases? How are you bringing the big tech folks to come and educate and train people on the ground?

Natalie, you want me to take that one?

Yes, please.

Okay, that's a great question. And thanks for being here. I think for us and the nonprofits that we're serving, we're kind of approaching this as we have to start the conversation with almost fluency in just the dialogue or the meaning of AI.

So we have a great AI for kind of intro course that we have available to all of the nonprofits in our community. That was actually designed before ChatGPT came out. So it's more than 600 days old.

And that one kind of level sets the conversation. It's an intro. I don't want to have a conversation about AI and have it only be about generative AI. I want to have a conversation about all the things AI.

So that kind of level sets just the language we use to talk about AI. And then really it's been about listening and onboarding and creating, like Anne said, those use cases. Like this is a specific use case. But if we skip that like foundational knowledge component and go right into the training, we're really doing a disservice.

And what we see is if we skip that part, they come back two months into using a generative tool and say, what's this? Because you didn't do the hard work first. So it's kind of that foundational knowledge piece, like what is AI in the folder that I shared in the chat.

One of the things that we've done with a lot of organizations is just start with how are you already using AI and you don't even know it.

And just starting like, so the language piece, the how are you using it before you even knew you were using it piece. And then we get in the use cases and let the use cases drive our training.

That kind of scaffold, I think, is where we're seeing people buy in overall.

Hopefully that's a good response. Thank you so much, Jodi.

Does anybody else want to touch on that before we move to another question?

Go for it, Anne. You just have to unmute yourself.

I'll just share one other little nugget, which is that we've seen a lot of success starting with the executive leadership team, which is oftentimes one person, but having them feeling comfortable and feeling like they have enough of background to make decisions on resource allocation has really helped with getting the rest of the organization into upskilling programs.

That that has been really important and really empowering. Thank you, Anne and Kavita, thank you so much for your question and thanks for being here tonight.

Our next question is from Jennifer Garcia, founder and CEO of Employee California. Jennifer, would you like to unmute yourself and ask your question to the panel?

Yeah. Thank you so much. Thank you, everyone, for being here. Wonderful, wonderful panel. I wanted to know what what out of the box low cost, of course, with nonprofits to get them started, tools that have been implemented that you could just name some some implementations, some practical uses that we could get nonprofits start on, started on right away.

Dupe, you're muted, but it sounds like you have something to say. And the microphone, there you go. OK, so I'll kick it off. And I, I, I love, love, love what you said, Jodi, about, you know, thinking about what people already use. Right.

So in our day to day, your teams may already be using Grammarly, right, for example. But there might be some shame around saying that they use Grammarly out in the open, which happens a lot, especially if it is a nonprofit organization where the culture is one that is, for example, academic. It might be there might be some shame around that.

I think bringing to the forefront things that people already may be using in their personal life, in their regular home life and seeing the examples come from leadership are actually a good way to start. Another one that I am a fan of, of course, is Otter. I mentioned Otter AI. I love this for meeting notes, especially when there is someone who is on the team, for example, who is neurodivergent and does not or may not choose to do two things all at once, which is the culture of a lot of organizations. I'll take notes and I'll listen at the same time. Right. That might not work. But encouraging that. Right.

I am a little biased, but I do believe that marketing teams should lead the way here. It helps in so many different ways with getting your marketing team up to speed. There are so many different tools that they could be using. I mentioned Otter AI. I mentioned Grammarly in addition to, you know, your chat GPT and so many different iterations looking at presentations. There's beautiful AI is another one that I enjoy using as a marketer. I think having your marketing team kind of bring those to the forefront and gently ease into, you know, you may have it where someone will say, wow, that presentation is really beautiful. How did you all get it together so quickly? And then letting them know afterwards that, oh, actually, we used X, Y and Z to make it happen. And then you'll find that, you know, your teams will be more comfortable with adopting once they see you using them in these cases to kind of elevate the work, if you will.

So those are just a few examples that, you know, that I have used that are very easy to use, low barrier of entry. And people, I would say, already are starting to feel very comfortable with using. And there isn't that much taboo around them. Thank you so much, Dupe.

Nathan, I think you also have something to contribute. You're muted.

Yes, yes, I figured that I had to hit two buttons at the same time. Apparently, that was too much for me. No, I absolutely agree with Dupe is that, you know, when you see you're really the advent of AI being not only, you know, affordable, but accessible. Now it's where, you know, I mean, truly, obviously, we're on an open AI forum. You know, ChachiBT is a transformative tool, whether it's free or paid for now. I mean, really, for nonprofit organizations and discounts for nonprofits to really transform their business in every way. You know, there's another way I'm a big fan. You mentioned Otter. There's another one called Firefly AI, which also transcribes very similar. And another one that I just recently picked up is called Play PL AI. It's a it's actually a device of a recording device that connects to open AI. And I was just in Geneva at the opening of the AI for Good summit. And I recorded our two hour long presentations with this little device that it's magnetic to the back of your phone. It's like a hundred dollars. And it transcribes. It sends everything to open AI transcribes. It gives you a memory map of the conversation, summarize it into agendas or next steps. And so I just love I mean, I'm just such an optimist on this. I love the accessibility and affordability of these tools to really level up nonprofits in almost every regard. I think the biggest limitation is our own creativity and and what we and how we can use those.

Woodrow, you mentioned you have something to share as well. Please unmute yourself.

The the there's a microphone icon in the bottom center. There you go. I press the button. Can you hear me? Yeah, great.

This I think is all really good advice. And I think a lot of the answer to this question tends to be around kind of how you can improve workflow productivity, which is all really helpful, in part because those tools are well developed and their commercial uses and and a good way to start learning how you can use these tool in your workplace. I would also encourage organizations to think about this from the perspective, in addition to what tools, what problem am I trying to solve? This is something that we set out. Our working group is specifically designed around problem statements first. And then technologists and companies and solution providers and developers and researchers, they can come and help you figure out what the right tool is or what needs to be written in order for you to solve that problem. And that's how we're going to get that kind of more interesting. That's next level nonprofit specific solutions developed. Is when we start from the problems that we're trying to solve. And that's where I think it really starts to get interesting. Thank you, Woodrow.

Jennifer, thank you so much for being here and thank you for your question. We'll see more of you soon. On to Jeremy Nguyen's question. Jeremy has been a part of our community since the very beginning. He's a fellow community manager at the Chan Zuckerberg Initiative. Jeremy, would you like to unmute yourself and ask a question? Jeremy, if he's still here. OK, I'm going to move on. But Jeremy, if you're still here, then you can go next.

Benjamin, would you like to unmute yourself and ask your question? Benjamin is actually going to be joining the OpenAI team soon as a forum ambassador. So you guys will be seeing him much more often. Maybe some of these folks have already dropped, but I like this question, so I'm going to ask it myself. Could you talk about how AI is helping with M&A efforts to demonstrate the impact of programs? So really, a big part of your work in the nonprofit sector is to demonstrate the metrics. Like what work have you done? How has it worked so that we can give you more funding? E.g. rapid experimentation, more robust analysis.

more creative signals to show impact. Does anybody want to take this one? I feel like this could be a Woodrow question.

Oh, and Woodrow.

Yeah, well, I'll start, because this is coming up a lot in our working group. We have a whole, one of the main themes that we saw in these problem statements was this, how do we get better at this? And so a great example of that is we're working with an international development organization that's been around for many, many decades, and we're partnering them with our own team and developers to look at their decades of reporting on their own programs and outcomes that are in many, many, many different and not easily compatible formats and make sense of what have we been doing that works best over the years to inform their own best practice. Getting that right also requires that good asset package that you can build on top of in order to contextualize those data and model various factors like political affiliation and language and a whole bunch of other things. But that's a really good example of where partnering those use cases with developers builds good tech that will have extensible use, both within the nonprofit sector. If we can solve that for those guys, it's gonna be helpful for others. But also I think other product development, if you get that right as a developer, you've built some good tech. Thank you, Woodrow.

Gail, would you like to add a little something?

Yeah, I completely agree with Woodrow. And I also, this is a personal belief, but while we need to work with existing nonprofits to train their staff and create new processes and procedures, I actually believe that startup nonprofits that start from scratch with an AI mindset will probably be more nimble and accelerate faster in this space. And there will be a lot of, in the for-profit world that merges and acquisition. I do think there's gonna be a huge shakeup in nonprofits. Even locally, we're seeing an economic downturn and smaller under-resourced nonprofits are going under. I see that as a natural evolution of things. And for the leaders who lead with AI and start fresh, I think they've actually got a lot of opportunity. Thank you so much, Gail.

Jodi, would you like to contribute?

Yeah, just really quick. I mean, I think, so I'll put this in the chat over here. One of the things I think that we're seeing, especially just with our organization and the nonprofits that we serve, we have so much information, right? So, and we don't have time and we have our own biases that we're looking at these documents with. So part of it is if we can, we're using this PDF summary thing in some circumstances to just get a handle on big picture, high-level things, high-level themes that are just sitting in our Google Drive. Like how can we move just even just select pages of things over, summarize those and get a perspective that we're not able to see. Sometimes we're too close to it to really see the forest or the trees on this stuff. And it gives us a different perspective. You're muted Natalie, the button's at the top.

Got it, yeah. It's kind of hard to tell who's still here. Okay, awesome. So the program manager for OpenAI for nonprofits, which was newly discussed in a blog post a few weeks ago is here, Alex Narwal. And I would love one, Alex, if you could please introduce yourself. And then after that, one of our grant recipients for the Democratic Inputs to AI program, Jai Wei, we know him as Peter. He has a great question about OpenAI. But first, Alex, could you introduce yourself?

Sure. Hi everybody, I'm Alex Narwal. I manage social impact partnerships at OpenAI on the global affairs team.

Awesome, so good to have you here, Alex. So Peter, are you still with us?

Okay. Peter also is joining us from halfway across the world. So I'm not surprised at all that some of these people have dropped because they literally gone on at 2 a.m. in the morning. Is that you, Peter?

Yeah, I got it.

Oh, Peter, do you want to ask your question about OpenAI and our plans for NGOs?

Yeah, sorry, I'm here in Taiwan and it is 7 a.m. in the morning. But I think this topic is worth for me to get up early and hop on this chat. And I'm very happy to share, I'm happy to ask that since we have discussed a lot of how is AI so potential to improve and better the use and better the development and the work of the non-profit and NGO, these kind of NGOs, these kind of non-profit organizations. So I wonder if OpenAI will offer any kind of helps or supporting planes or these kind of programs to support these kind of further AI use in a non-profit or NGO organizations like that. Thank you.

Thank you, Peter. Awesome. So happy to answer that. OpenAI launched a initiative about three weeks ago called OpenAI for Non-profits. The major initial thing this includes is a discount for ChatGPT team and ChatGPT enterprise. And we also have a small help page that answers some questions. We'll be developing more resources and content going forward. So we hope to have some kind of like getting started guides for non-profits and things like that. I've gotten a lot of ideas from you all today about things you could include there and we'd love to hear more. So keep an eye out for that over the next couple of months or so. One other way we partner with non-profits is through accelerator programs. So often we will work with an accelerator around either a particular theme or a particular challenge and we'll bring a bunch of non-profits into that accelerator. They could be at different stages. Some of them might be kind of at an R&D stage or some of them might be trying to scale and we'll help them develop their product. OpenAI brings our best solutions engineers to help with that. So we have a couple ongoing now and you can look out for more to be announced soon. And if you see any cool non-profits or you know any cool non-profits or you're part of one that is developing an AI solution, then they should definitely consider applying for those accelerators. And last thing I'll say on that is on Tuesday, we actually have the demo day for our most recent non-profit accelerator, which is in partnership with turn.io. That's virtual and open to the public. So you can all check that out if you're interested and see what a number of kind of smaller, innovative non-profits have been working on. Thank you so much, Alex. And we hope to see more of you in the forum.

And also just for those of you who don't know, if you're not super familiar yet with ChatGPT+, it now includes DALI. So if you work in the realm of marketing, such as DuPay, then you can also now start to incorporate multimodal solutions. You can upload images and work those into your marketing and fundraising methodologies. But more of that to come.

Oh, and Alex just dropped a link to demo day. Thank you so much, Alex. So I actually think we did a really wonderful time, like sharing a lot of information. We got to-

all of our very highly upvoted questions. A couple of logistics guys, at the end of this event, when you go to your messages tab in the OpenAI forum, and if you don't have the forum already bookmarked, please do so that you can see when you have notifications. People that attended tonight might be trying to reach out and collaborate with you. If you go to your messages in the wake of this event, our entire chat will be there. You can find people that have asked questions and reach out to them and DM them. You can find links that have been posted here.

Any of our speakers present today, Allison, Gail, Ann, Dupe, Woodrow, Scott, Nathan. You guys can respond to questions that we didn't get to live if you like and we can keep the conversation going. Please observe that in the wake of this event.

We're going to move into, I'm going to share a few updates with you guys about in-person events that we have in the very near future. Then I think if you need a bio break, if not, go ahead and jump into the networking event that we're about to host for you. The networking events, all you'll have to do is follow the link that Laura is going to post for us here, and you'll be randomly paired with somebody else who's in attendance tonight for just three minutes. We have some prompt questions for you to help break the ice, if you're interested in doing that. If that sounds terrible, no problem. I know virtual networking can be really anxiety-producing for some folks. If that is not your jam, no pressure, but it is there for you.

Laura, if you could please give us the notification for the one-on-one matching. In the meantime, I'm going to share some updates with you all.

As most of you know, the folks who have been here for a long time, we started this community as a means of connecting with external stakeholders that had expertise that we do not have at OpenAI to support our research programs. This is still very important to us. Representation matters when we're collecting data, when we're annotating data, when we're evaluating our models. We turn to the OpenAI forum and all of you, the experts, whether expert practitioners in your domain of expertise, experts because you've been studying, researching, producing a body of knowledge as a PhD for a long time. We turn to you to support our data annotation and even write evaluations to make sure that our models are safe and objectively correct.

In order to honor you, in order to make you feel like you are a part of our team as you are, we're inviting people who have already trained with us and collaborated on any research project in the past year to hang out with us next Thursday at OpenAI. The spots are limited, but we still do have space and you can register online in the forum. If you can't make it in person, we're hosting this virtually in early July. Because there's more space for that one because it's virtual, if you haven't already participated in research with us, but you're curious about it and you'd like to learn from others who've done it, please join the virtual events.

We actually have somebody here tonight. He's in Texas, it's getting late, but Naeem, are you still here? Naeem is a computer science graduate and he's in participating. Is Naeem here? There he is. Hi, Naeem. Hi, I'm here. Naeem is going to be saying a few words. He's going to talk about his experience in our virtual mixer in a couple of weeks. You can ask Naeem questions about what it's like to actually be contributing to OpenAI Researcher as an external stakeholder. He's also a very kind, loving human. I hope you get to meet him along with many other really awesome people in the OpenAI forum. Come meet them virtually or in person.

Thank you, Natalie. I really appreciated it. It's been a fantastic time working with OpenAI and looking to develop the models and stuff that we're training at OpenAI. It's been a great time working with the community. I've been connected with a lot of really intelligent, smart people along the way. It's really challenged me to think deeper about how AI is developing and how the part that we play into it. It's been a fantastic time. I'm really looking forward to the virtual mixer that we'll be having here in a couple of weeks. But the team has been fantastic. The platform and the ways that we're looking to connect with one another and making sure that we have one-on-one opportunities. We're really looking for the right questions and research whenever we're trying to implement new knowledge into our models has been amazing. I really hope you guys show some interest or have an opportunity to reach out and work with OpenAI and look for new ways to help out.

Thank you so much, Naeem, for holding it down while I lost Internet access. The last event I just wanted to share with you guys is in July. It's in person. You are all welcome to attend. We'll be hosting the lead astronomer for the Event Horizon Telescope, who has recently published some of the most innovative and detailed photos of the black hole. He is going to be discussing how we can leverage AI to support scientific research and explore new realms that we haven't even begun to explore. So I hope you guys can meet us for that. I really love the in-person events. That's why I'm spending so much time talking about them tonight.

Thank you all for joining us, especially our nonprofit leaders and the panelists today.

Thank you so much, guys. It's truly an honor to have you here. Thank you for being the torchbearers and forging space in the nonprofit community. Gail, that was really cute. I like that.

And if you want to meet each other in a one-on-one network, but like matching, dating game, please, you can select the Join Now button up at the top. And other than that, I will see you all very soon. Thank you so much for joining us tonight.

And Nathan, thank you so much for helping coordinate all of this. This has been six months, seven months, maybe nine of work in the making. Thank you so much. Truly an honor. It was so great to be here.

Thank you, Natalie, and for everyone that attended and our panelists. Thank you.