Episode notes:

• Follow Shannon on LinkedIn

• Learn more about Fast Forward 

  • Check out their Accelerator Program

  • Sign up for their Weekly Resource Roundup

• Access the Google.org Impact Summit on demand

• Explore the tech nonprofits: the Khan Academy and Wikimedia

• Learn about the Google Generative AI Accelerator for nonprofits

• Check out what Reboot Rx is doing to fast-track the development of affordable cancer treatments

• Learn more about Jared Chung and his work at CareerVillage.org

• Dig into Google’s Get Back Time Campaign

• Learn about the Google Fellowship from Jen Carter

Transcript:

This transcript was automatically generated.

Welcome everyone to the Impact Audio podcast, brought to you by your friends at Submittable. I'm your host Sam Kaplan. Earlier this month, I got to attend the Google Impact Summit in Sunnyvale, California, an event all about how tech and especially AI fits into social impact. Shannon Farley, who is the cofounder and executive director of Fast Forward was also there, and she hosted a fantastic session titled Funding AI Responsibly that discussed the nuances of funding AI powered nonprofits. And today, she joins me to chat about what she learned from the session she conducted and from this year's summit as a whole.

Shanna, and welcome to the Impact Audio podcast. I'm super excited to have you join me.

Hi, Sam. Thank you so much for having me. It's a real honor.

Yeah. Well, thank you. And before we get too deep into our experiences at the Google Impact Summit, like, start by telling me a little bit about you and your work and, like, what Fast Forward is all about.

Yeah. Thank you for that. I'm the cofounder and executive director of FastForward.

We started FastForward over ten years ago because we were solving a problem that I had had as an early tech nonprofit founder. I was looking for a community of people who understood that tech could be a force for good. I was looking for context on how to start and scale powerful nonprofits that had tech at the core of the impact model, and I needed access to funding.

And so we started with an accelerator program that's had over a hundred organizations go through it. And over the years, FastForward has scaled. Now we do a lot to bridge the nonprofit and tech worlds to ensure that we're getting the best of technologies in communities that need it most.

That's really cool. And, like, I think that a lot of people listening to us, like, may not be super familiar with, like, what exactly makes a nonprofit, like, tech focused. So, like, let's talk about this kind of little corner of the nonprofit sector. Like, tell me about, like, what is a a nonprofit that has a technology focus, and, like, why do they exist? And, like, why would why would an organization form as a nonprofit as opposed to something like a social enterprise?

Totally. Yeah. We are a small but mighty corner of the social impact world. There are about a thousand tech nonprofits globally. What that means is they are organizations that have tech at the core of their impact model. So think Khan Academy. Think Wikipedia.

These are organizations where the work is the technology as opposed to what we would call a brick and mortar organization that uses tech well. And why it's important to have nonprofit business models is because if you decide to serve the beneficiary, actually, everything about your organization would be different than if you were a for profit. You're building for the hardest to reach customers and the hardest to reach communities.

You can leverage open source tools and in kind donations from tech companies. You have a different impact model and business model.

So we basically have, over the last ten years, written the book on what it takes to start and scale tech nonprofits, and we've seen some really impressive examples. And many of them were shining, during the Google Impact Summit through the Google Generative AI Accelerator.

Yeah. And I'm glad you mentioned that. We're get we're gonna dive right into the Google Impact Summit. We were both there, and this was a summit that really, like, focused almost entirely on AI and the emergence of this technology and how it is beginning to, like, reshape not only the work that these thousand tech oriented nonprofits do, but it's beginning to, like, create, like, a whole new category called AI for humanity.

You know? And this is, like, artificial intelligence and service of solving some of these, like, wildly complex and intractable problems that the nonprofit sector has been attempting to tackle for a long time. Like, so before we jump into, like, the Impact Summit, tell me just a little bit about, like, artificial intelligence and kind of how FastForward is approaching, you know, this new technology and the way that, the nonprofits that you work with are beginning to integrate it.

Yeah.

That's one of the things that's, I think, exciting about AI.

Many organizations that we work with have been using early forms of AI, you know, for the last eight years, early chatbots, early machine learning models. What we've seen now is with the release of newer models, cheaper models, we're seeing lots of folks adopt it and have meaningful impact through AI. A classic example for us is Reboot Rx.

RebootRx was started by researchers from MIT and Dana Farber who all had loved ones who had experienced devastating cancer diagnoses.

And the devastation was not just to their health, but also to the financial well-being of their families.

And so they started, RebootRx seven years ago because they wanted to use machine learning and AI to search through hundreds of thousands of research papers to surface generic drugs that could be used in the treatment of cancer.

Why it's a technological breakthrough is because cancer research has been happening in nonprofits for a long time.

By using technology, they can do it cheaper, they can do it faster, and they can surface there's no business model for generic drugs. So they can surface these drugs so that low income patients get access to them immediately.

And to your earlier question, like, when we talk about AI for cancer research, that's sort of a classic for profit AI trope. There are a lot of biotechs that are building technology to discover AI technology, to discover new drugs.

With RebootRx, they're surfacing generic drugs. New drugs are expensive. Folks are not gonna be low income folks will not be able to use them for, you you know, up to ten years. With generic drugs, they can use them immediately.

Yeah. There's something that's just kind of magical about combining artificial intelligence with the people who are part of the nonprofit community and the nonprofit sector because, like, nonprofits just think about the world in a different way. Like, I mean, you know, historically, like, they're really scrappy and they're really innovative and they get more done with less, but they're also approaching these, like, really, really challenging problems, like, you know, finding new, you know, cures for diseases and, you know, helping with, like, the refugee experience. And and they're just tackling these types of issues that, like, for profit corporations, like, may or may may not be that interested in in dealing with. Right?

Yeah. There's no profit model for reaching these audiences.

The thing that heartens me about this moment is that technology has always been at the core of social justice movements. If you think about the civil rights movement in the sixties and seventies, phone trees, were key to organizing. When you think about human rights movements around the world from air spring, social media platforms and mobile phones are key to how folks organize. So when I think about AI for humanity, I really think about this as the tool for unlocking the kind of impact we've always wanted, but the technology is so much better, so we can do it much, much faster.

Yeah. Alright. So let's dive into the the Impact Summit. So we were both at the Google Impact Summit. Yeah.

It was fun. Right?

It was totally fun. It was so inspiring for me. And, I just love to hear, like, what are your high level thoughts about the event and, like, you know, what did you learn about the future of of nonprofit and AI?

Yeah. There were two, like, huge themes for me. One is that there were these very impressive examples of AI powered nonprofits that have been playing with these tools, deploying these tools, and impacting humans in really wonderful ways.

We got to see Jared Chung from Coach, pitch on stage. Coach, Fast Forward, has supported Career Village over the years. Coach is their AI enabled coach that's helping young people access career advice.

The thing that was so amazing was to see like, this is an organization that has been around for, like, eight years. And by all accounts, it's hugely impactful. They have a question and answer site where people can get information on what to do and how to get a career. Most kids get, like, eight minutes with a college counselor.

Career village has solve has solved that problem. What it they've been able to do is unlock this power trove of data that they can use in their AI models to have, like, a real coach where they can capture information and get nudges and see how kids do after the advice, which is revolutionary in the career advice space. It's so exciting to see them pitch.

So the there are these great AI powered nonprofits.

The other thing that I was is so thrilled that Google dot org is doing is their announcement of the get time back campaign.

This is a campaign where any nonprofit can leverage AI tools to make their organizations more efficient. Think of using Gemini to rewrite grant reports. Think of using smart calendaring to free up your staff's time so they can focus on things that are more important. Like, as someone who cares deeply about making sure that nonprofit workers' lives are fulfilled and delightful and well paid, the idea that we could take off some of this administrative burden feels like a step change in the sector, and it's really exciting to see.

Yeah. I totally agree with you. Like and, so I was in your session. It was titled funding AI responsibly, and you talked about, Career Village and and Jared. One thing that Jared said about, like, the development of the AI functionality is that, like, as an organization, they didn't approach this with, like, an AI first mentality. They approached it in terms of thinking about what technology is going to be best for this particular use case, and it happened to be AI. Like, do you sense that more and more, like, AI is going to be, like, the right technology to help scale these, you know, various technology initiatives happening in all these nonprofits?

Yes and no. Right? Like, AI doesn't fit every organization to build on your own. You have to have technical chops on staff.

You have to have financial resources to be able to deploy it. So there certainly is a subset of organizations that cannot build their own, but they can leverage it in their day to day work. And that's what's cool about the get time back campaign. It surfaces all of those incredible resources.

What is true about tech nonprofits and always has been is that and it's why it's fun to work with them, honestly, is that they're not precious about the type of technology.

Their North Star is impact. So if you could get there with, like, a pretty good chatbot, you wouldn't need to build a fully baked coach.

Right?

For Career Village, they wanted much more nuanced answers. They wanted to be able to have feedback between student and coach. They wanted to be able to do interview prep questions and review resumes. Like, they wanted all of these things that you would get if you had you could afford, quite frankly, a great career coach.

That is possible when you build your own AI platform.

And, so it made sense for them, but they weren't to your point, they didn't start out with this is the only way forward. It just ended up being the best way forward. And the good news is that it's working.

Yeah. For sure. And you mentioned, like, the announcement, from from Google about supporting nonprofit organizations.

And we've seen, like, similar announcements from other big tech providers out there. Yeah. Like, do you feel like this is kind of the right time to be a technology focused nonprofit that wants to integrate I AI in your work? Like, in addition to all the amazing services that FastForward can provide these tech startups, like, now they have additional support from Google and from Microsoft and, you know, for many other companies out there. Like, like, do you feel like this is sort of a, you know, an inflection point if you're a tech nonprofit?

I do think it's an inflection point for tech nonprofits.

One of the things we learned during COVID was that tech nonprofits were really ready for a moment where we transitioned to a mostly digital and then hybrid world. They had products in use. People were playing with them. They had business models. They had figured out a lot of things. And as a result, they scaled their impact exponentially.

Tech nonprofits are ready for this moment. They've been training for it in so many ways. And so I think it's great to see Google and Microsoft and Salesforce and others come to the table with technical talent and funding and product, because it actually it was gonna take all three of those. AI is still early days. And even as tech nonprofits are ready to build, deploy their own AI first products, the technologies in their consumer interface is pretty new. So we need to be a holistic tech community of platform companies and nonprofits and CSR leaders, honestly, to figure out how we work together to make these things happen.

Yeah. I totally agree with that. Like, in in a lot of the research that that I and my colleagues have been doing around, AI adoption in the nonprofit sector, like, the area of responsible and ethical AI continually comes up. And, you know, clearly, I think, and rightly so, you know, there is concern that AI has the potential to do harm and that, you know, you know, algorithms can introduce bias into, you know, various products. And and I'm just curious, like, how how are you, and how is fast forward, like, sort of thinking about this area of, like, responsible and ethical AI in light of this, like, sort of rapid adoption that we're beginning to see?

Yeah. I think Jared from CareerVillage's answers were great on that one in particular. Like, one of the things we're seeing groups like Career Village do is lead with ethics, lead with ensuring the data is unbiased because that's their mission. Their mission is to serve the hardest to reach child in the hardest to reach community, and those kids have been excluded from so many other services.

They don't want that exclusion to have a through line to the AI product they're building. So it actually took a long time to launch Coach. Right? Because they wanted to make sure it was safe.

And they worked with community partners really closely to see how the kids were using it to if there were any hiccups, they wanted to address it right away and have immediate feedback to the product team at CareerVillage for how they could fix it.

A for profit company wouldn't have the privilege of doing that because they don't have the same goals.

Right? They're chasing speed and first to market, not necessarily safety of users.

And so this is a space in which I think nonprofits have a lot to teach platform companies and other for profit startups, of how we build in ethics, safety, and humanity into the products we're building. And it Yeah. It's great that they work together.

Right. Jerry just did a fantastic job of describing, like, this sort of, like, you know, very careful, methodical, well thought out process where, like, he incorporated so many stakeholders, including, like, kids who are using the platform, you know, into, like, not just helping from, like, a, you know, UX perspective, but really, like, looking at the output of things and tailoring and just, you know, all of the effort to make sure that they got it right before this, like, got rolled out to a larger audience. And I totally agree with you, Shannon. Like, you know, this is something that the nonprofit sector can totally lead on.

And we hear so much about how, you know, people regularly say, like, oh, nonprofits should operate more like for profit businesses. And I'm sure their heart's in the right place when they say things like that. They're thinking about, you know, like, you know, project management and, you know, efficiency. Right?

But in reality, I think to your point, like, nonprofits have a lot to teach the for profit sector as well, especially around AI adoption.

Yeah. And, you know, I don't know a nonprofit founder in the world who doesn't want their operations to be more efficient, but often that comes at quite a cost. I mean, that's one of the things that's really exciting about this moment and the get time back campaign is that there are all these resources on how to be more efficient so that we can spend more of our energy and financial capital on supporting our beneficiaries.

So as someone who spent my career in the nonprofit sector, I'm very excited for what is gonna bubble up.

Like No.

And I I also wanna give credit to there's a number of tech companies that have funded tech nonprofits to think about how you infuse ethics and infuse safety into what you're building. Like, the Salesforce AI Impact Accelerator did a really good job of that and ensuring there were opportunities for other tech nonprofits to convene and in safe spaces, figure out how to do these things well.

Google has always put fellows into tech nonprofits so that you can work in step with Google engineers who've seen it in their jobs and nonprofits who are seeing it in their jobs. So, like, we're infusing best practices throughout the sector.

Yeah. Jen Carter from Google dot org, did an episode with us, and and we talked a lot about that fellowship program. And, really, really how it differs in in the way that they are able to take Google staff and put them on a project with nonprofits for, like, a really long time. Like, it's not like a let's all get together for an afternoon and do a hackathon.

This is like, let's take a long time and really, like, build something significant for a nonprofit organization.

Like, that's pretty incredible. Right?

Yeah. And generous.

For sure.

Particularly, they're lending out AI engineers. Right? And I think they're doing it because there is a business case to being thoughtful about how we build for people who have never been at the center of our product road maps.

That's the most marginalized, the most excluded. And so, yeah, it's a it's a wonderful project they have.

Yeah. And so let's circle back to fast forward a little bit. Okay? So, like, I am I'm listening to this podcast, and and I'm a a nonprofit leader who has a tech oriented nonprofit.

And I am super excited now because I wanna work with fast forward. Like, help me understand, like, the process for all of these, you know, burgeoning tech nonprofits who might want to, be affiliated, with your organization.

Thank you for that question. There's lots of ways to get involved with us. Cool. First and foremost, we still run an accelerator program. It's a thirteen week program where we give emergent tech nonprofits money and training and love to get on their feet to build great things. So folks should apply to that and learn more about that. We also have a number of resources, including a newsletter called AI for Humanity, which features AI powered nonprofits, AI powered funders, folks who are thinking about how these new emergent technologies can be leveraged for good.

And then, actually, our most beloved product is something called the weekly resource roundup. This is a weekly email of fundraising resources, marketing opportunities, and prizes.

You would be surprised at how many thousands of people read it every week. I'm often stopped at conferences or even at an airport line last week when someone told me they got funding off of that email, which is why we do the work. Yeah.

That that is really cool. That's a great outcome of, like, you know, one of the many, many things that that your organization does. So, like, given that that, the topic at the Google Impact Summit was, like, funding AI responsibly, like, what's the biggest or the best piece of advice that you would have for funders who want to start funding AI powered nonprofits?

Yeah. The thing that's interesting about it is I don't think the questions you would ask are actually that different than the questions you would ask a tech nonprofit.

You wanna understand why this technology, why AI, what's the analog, and is AI actually gonna be better? You wanna understand who's on the team. Do they have folks on and talent on the team to build it internally? Because you need that.

And you certainly need people who can manage contractors if you're not building it internally. So you wanna understand that. You wanna understand how folks are thinking about ethics, how they're thinking about their data, privacy and security, and removing bias.

And you AI is still expensive, so you wanna know the business case for it.

Yeah. And I think finally that that the upside is bigger than the downside.

This space is experimental.

Anytime there are live experiments, there's gonna be mistakes.

So you wanna know that you're the impact you're gonna have is tenfold better. That could mean ten times more people are reached, or in the case of coach, the quality of the intervention is so great that it justifies the staff investment, the human capital, and the reputation risk to do it. I don't think there's that much reputation risk at the moment, so it feels like most of the world is going all in on AI. But it's important to think about when you are funding it. It's important to think about when you're building it and when you're using it.

Yeah. I I completely agree with you. It also feels like there's a big opportunity right now. Like, a lot of funders are beginning to think about movements like trust based philanthropy, like, that introduce more equity in that, you know, nonprofit funder relationship and, like, funding, nonprofits who are led by people of color and that serve marginalized communities.

You know, it just feels like it's a really good opportunity right now to start thinking about, like, those kinds of principles and practices in association with funding AI based, projects and and proposals. Right? So, like, if you're funding AI, it feels like it's going to be really important to like provide a general operating support type of grant rather than being really prescriptive about the very specific outputs or outcomes that you're expecting from a nonprofit organization. Like, do you think that this might be also an inflection point in terms of, like, the larger kind of, like, relationship between, philanthropies and nonprofit organizations? Like, a chance to take a big step back and think rethink about how that funding process works?

Yeah. I hope so. Right? Like, everything we're trying right now is experimental. The technology is experimental.

Even though it's been in operation in Google land for, like, twenty years, in the social sector, it's still experimental. So you need funders that have that trust based relationship that are willing to give you the pilot money to just build something and see if it works, and then be okay if you choose it didn't work for any number of good reasons. Right?

What we wanna avoid, I think, as a society, is no more, like, shiny object funding. We wanna fund real learning, real growth, and and impact.

So that's the opportunity, and it really only works when philanthropists are in deep communication with nonprofit leaders and builders.

I think there's also an opportunity for philanthropists to put it on themselves to sit in, like, the learner seat.

There's an underinvestment of technology in philanthropy.

So one of the things we recommend to new AI funders is to run AI experiments internally.

Play with these new technologies, build your own bots, see what it takes, build the empathetic muscles required to advise your grantees appropriately.

This innovation is not sequestered to builders within nonprofits. We, as funders, should also be thinking about how we leverage these resources to make sure that more money is going into community rather than our operations.

Yeah. It makes perfect sense to me. And, like, you know, on the stage, we heard Google employees talking about, like, the function of philanthropy in part is to be risk capital.

And, you know, sadly, like, historically, not a lot of foundations and philanthropies have really treated their capital as being, like, you know, okay to, you know, to to to to fund projects or initiatives where there is some risk involved. And I think that's exactly where we are when it comes to artificial intelligence and the nonprofit sector is like there is going to be risk. There is going to be failures. We're going to have to learn as we go along the way. Like, this is the time to really, like, sort of double down on that idea that, like, philanthropy should be funding risk and we should be doing some experiments. And, like, let's figure out what doesn't work and what does work, and then let's find ways to scale the most successful AI based projects out there.

Yeah. That's exactly the kind of advice we give new tech philanthropists. Like, one of the things I say is the money has been spent. You put it in a DAF. Like, you should see what you can do with it to take your expertise and your talents, and your capital and put it to work. And that includes being a little bit more risky in how you deploy the capital.

Right. And so, what about, like, on the other side? So if you're a corporation and you are wanting to, get engaged from a funding perspective with with these nonprofits who are tech focused, like, they can work with your organization to do that funding. Right?

Yeah. We work with a lot of tech companies who are interested in investing in a portfolio of organizations that are aligned with their issue areas. So if you're a foundation or CSR leader that really cares about education in India, we have a portfolio of organizations that you can support. If you care exclusively about STEM education, we have a group of organizations you can support. We have about a hundred organizations working across issue area in every imaginable ge imaginable geography.

Very cool. Alright, my friend. Anything else? Like, what what didn't we cover about the Impact Summit that that you thought was pretty amazing that that we should end this, podcast on?

I think what is so special about what the Google Impact Summit did in this moment of time is inspire us to move forward now.

Like, the technologies are advanced enough that they can be deployed in our community.

There is enough capital to invest in these organizations, and there are builders who are equipped both with proximate experience to the folks they wanna serve and tech talent to execute against those visions.

So there's a the time is now to build something that matters, that can change people's lives.

One of the things I'm worried about as a human on the planet is that these technologies are moving so fast that if we as a social sector do not intervene, we may be left behind.

The divide between who has access to helpful things and those who do not is gonna widen if people who aren't caring about those in the last mile aren't building for them. And we have this really special opportunity and resources to start to solve those problems today. And I hope that's what we all take away.

I definitely took that away. So, Shannon, this has been a fantastic conversation. We're super excited that you're able to join us.

Your your organization, like, develops tons of great resources for nonprofits. We're gonna put links, in the show notes for everyone to check out. So, I think that's all I got for you. Any any closing words of wisdom for us?

No. Thank you for having us. And I I hope you come and check us out on our Substack at AI for Humanity or find us on our website at f f w d dot org. We're here to continue these conversations and build exciting tools with you. Thank you.

Thanks for tuning in to Impact Audio produced by your friends at Submittable. Until next time.