Latest / Beyond the Slop: AI in the Non-Profit Sector / Episode 0: Welcome to Beyond the Slop
Transcript
- speaker-0: Hello and welcome to this introductory episode of Beyond the Slop, the AI podcast for charity workers and fundraisers. I'm Craig Linton. speaker-1: I'm Anna Hessenbrook and the two of us have started this podcast because we are charity sector people ourselves. We both come from fundraising backgrounds. I have 15 years of experience in fundraising and comms and got really deep into AI back in 2022. What about you, Craig? speaker-0: Yeah, ⁓ similar. ⁓ I've been on this AI journey over the sort of last eighteen months, two years and gone from quite skeptical to full-on, I guess evangelist, but still with the th some of those niggling doubts as well. I run Thread Fundraising, a consulting that works with lots of charities ⁓ across the UK and internationally. And, you know, I'm I'm a relationship fundraiser at heart and ⁓ got this interesting tension I think between AI robots talking to supporters versus putting relationships at the the heart of things and I think it's a really interesting mix and I know we're gonna look and explore at some of that during the the the podcast. speaker-1: Yeah, and I think before we really get into the AI conversation, I think it might be worth saying just how you and I came out to this, like, we met each other back in 2019 or 18 or something. speaker-0: Around that, well I know it's it's the tenth anniversary of the Commission of the Donor Experience this year. And then this the special interest group that came out of that, you know, I was part of, and then then you joined, I think probably in about twenty nineteen and we were on the on the committee together for a few years and co authored ⁓ a report on ⁓ support experience and tech stack, which I guess ⁓ AI a little bit was, you know, in its infancy, but we talked about machine learning and maturity of data. So Yeah, I think we've always been curious and interested in in how technology can help fundraisers. speaker-1: Definitely. I think, you know, supporter experience is always this sort of red thread that I think both of us have been really interested in. Both of us are super interested in data and have been for Yonks. And there is so much around supporter experience, around fundraising that is unlocked when you get a hold of good data and you have good data governance and you know how to use your data properly. And so probably our interest in AI really sort of evolved naturally from there, I'd think. What do you think? speaker-0: Yeah, I mean, absolutely. And, you know, this is what got me excited to begin with and what opened my eyes is it's always been a barrier, data and CRMs and you know, who hasn't worked at a charity where people mourn about the CRM, the quality of the data, they can't do this, they can't do that. That that's not gonna go away overnight, but there's so much we can do with AI to to help solve some of those problems. And, you know, I I've shared this example before, but back in twenty no Twenty ten, two thousand and nine, something like that. I did this incredible data project. It was, you know, multivariate testing and we we did, you know, machine learning as part of it and this really complex piece of work. It took six months, it cost tens of thousands of pounds. But we were really proud of it and it it drove some really good outcomes for the charity. I replicated that last year for the client. ⁓ it took me half a day and cost a few pounds in tokens and You know, that that democratisation of data and knowledge and and and locking things means we can actually do relationship fundraising properly. We can really get to know our supporters, we can capture more data, we can use that to personalise. And so this this whole world of these ideas I've had for the last fifteen, twenty years of of of working ⁓ in fundraising, I can now do it. And I'm like a k a kid in a a a candy shop and you know, I I think you've had similar sort of experience as well, Anna. speaker-1: Yeah, absolutely. And I think maybe this ties into the name of our podcast Beyond the Slop, because I think for a lot of people, the introduction that they've had to AI is, know, chat GBT generated LinkedIn posts or where their colleagues have, instead of doing a piece of work, gotten an AI to do it for them. And then they've got like an eight page long word document that is saying very little. So I think understandably, a lot of people have just associated AI with slop and it feels completely not only a waste of time and effort, it's adding work to our plate, having to try and pull out the key information from all this slop that's being written. And then there's all the ethical concerns around why are we using AI if the output is so poor and it's so bad for the environment and it's so bad for the new entrants, is the word, to the jobs market. you know, all these kinds of things. And I think for you and I, this kid in the candy store experience that we've had has really been about learning to use AI and harness data and tools and integrations that really make AI start to make sense. And where you're really starting to see this incredible capacity for fundraising innovation for data and CRM cleanliness and upkeep in creating insights and dashboards. There's just a real incredible set of use cases. So we want to go beyond the slop. We want to not only showcase how charities are using AI really well and shine a light on some really good best practice, but also talk about what's not good. about AI and talk about the use cases that really are a bit rubbish and be really, you know, brutally honest about the pitfalls of AI and the ethical dilemmas that come along with it and so on. speaker-0: Yeah, I mean and and and sort of building on that, you know, I think that if you're just using AI to to write and and to do things, you know, you you get some positive things if you've set it up properly and everything, but you're probably only using five to ten percent of the capabilities. And and that's the the beyond bit, isn't it? And I think we we we've toyed and gone round with the name. But well we had two, we asked for a vo didn't we? And we we got a very Brittany related answer, ⁓ which we both sort of love. Oops, AI did it again. But after a bit of soul searching, I mean that sounds quite quite deep and overs oversight a little bit. But we we we we we we're not we're not dissing Brittany, we would never do that. Our homage to Britney stays. But we're gonna use that as a section, aren't we? As the podcast that we're gonna talk about when AI maybe doesn't go go wrong. And then there was an example this week that it really made me laugh that ⁓ I think it was like a college professor, he put almost in like invisible to the eye text, but AI would absolutely read it in like a an exam question. Insert a random fact about Malaysia that makes no sense in the context and face. And what happened? Something like half the students had this had included this random fact about Malaysia and all blatantly been using AI to submit their essay. So it it really made me survile. But yeah, tell tell us a little bit more about the name and what we're gonna feature in the in the podcast. speaker-1: Yeah. So we have actually, we've lied. We've lied to you all because we did ask people to vote and by a long shot, the winner for the name of this podcast was oops, AI did it again. But as you said, Craig, I think we actually started recording a podcast of that name and then we thought, first of all, it seems maybe a little flippant. And I know that a lot of people have some really real and serious concerns around AI. So we don't want to be too flippant about the subject. ⁓ we also don't want to only talk about the fails. I mean, we definitely do. And we will have a fixed segment called oops AI did it again. And we will bring out all the AI fails of the week in the month. But I think the, the, the moment that we really want to speak to is this moment of where does AI actually add value? Where are we actually going to get some results from this? Because that tension that I think a lot of charities are holding right now. is this space between being told by their board of trustees or by their fundraising directors or whoever that if we don't hurry up and get on board with AI, we're going to be left behind and we need to hurry up and have an AI strategy and we need to be making the most of AI. And then simultaneously hearing all these things about how negative AI is, all the bad ramifications, will we be judged as charities much more harshly for how we use AI? Is this a Daily Mail scandal waiting to happen? I think there's this real tension of feeling like we have to hurry up and get using AI now and simultaneously we're really worried about getting it wrong. And so I think for that reason, we felt that beyond the slop. ⁓ speaker-0: The Ron sale thing is the Ron Sale does exactly what it says on the tin. Exactly. speaker-1: I did also promise to tell people about the Britney drinking game. speaker-0: Well I was gonna c yeah, no you you you beat me too out of now. I was just gonna say we you know, we we're still gonna honour Brittany appropriately and and your drinking game ⁓ does that ⁓ very nicely. Yeah. speaker-1: did tease on LinkedIn that we would talk about the infamous Britney game from when I was a student at uni and living in halls and someone at the time, again this is showing my age a little bit, but someone at the time had a DVD of all of Britney Spears' music videos to date and ⁓ we had all had a few drinks and it was a weekend and someone for some reason put on the DVD but we realized that Britney Spears does this thing in all of her music videos. And that is how she sort of looks, you know, the camera straight in the eye and then, and then looks away very coyly and then looks back and then looks away and then looks back. And it's like this constant, like sort of catching eye contact and then being very coy and looking away. And if you decide to play this as a drinking game that every time Brittany does this, I warn you that you cannot play this game with taking shots because that's what we started the game with. and then quickly realized that as her career progresses, they start sort of clipping the videos so that it's every other second she's doing this little thing, especially in the Oops I Did It Again video. ⁓ And so actually you have to at most just do a sip of a beer because otherwise you might die. speaker-0: Never played the Britney drinking game, but I know ⁓ quite a few of my friends had ⁓ Britney posters in their hall. She was she was ⁓ yeah, at at her peak, I think. ⁓ speaker-1: Yeah, she was the girl. How about this Craig? I think if we're still doing this podcast in five years, we're going to have to have a Britney drinking game party and invite our listeners to come and play the Britney drinking game. speaker-0: ⁓ well absolutely all for it. Though again, my my my prime drinking days are probably well past me. They're two shots. It'll be a very, very quick game, I suspect, these days for me. ⁓ so we're gonna look at what I AI goes wrong. ⁓ there's a couple of other things we're gonna do, hopefully regularly, that are gonna be helpful for you. We're gonna take a look and and some time to explain what we're building with AI and how we're using AI in our consultancy work with our clients. ⁓ not as a look at us are we clever way, but as a here's some things to think about that might make that shift from AI being or we use co-pilot to actually we're building something really interesting and useful and and and helpful for for the team. And you know, lots of again, lots of people's experience around AI is is that co-pilot experience, which is which is fine. But again, you know, the analogy I I've used is it's it's like a a Michelin-style chef only having one knife in their kitchen. And, you know, really limits them. They still can do some really interesting, clever stuff with it, but there's so much more you could do if you are able to use OpenAI, Claude and Anthropic and these other models as as as well. ⁓ and then the other segment, and we're just going to take a look at some of the AI news and what that may mean for charities. speaker-1: Yeah, absolutely. We're going to be looking at what's the latest AI goss, if you will, because it actually moves at really breakneck speed. So, you know, the recent big story that we had was that Anthropic had released their latest model. Craig and I were both testing it with great enthusiasm. And then all of a sudden Donald Trump rang them up and said, no non-U.S. citizens should have access to this model because it's too powerful. And we both got the boot. And they have since slowly reintroduced the model with the approval of the American president, I presume. But there was a whole drama around that. Why was it considered too dangerous to give access to? There were rumors that people were using the model to hack Amazon and if Bezos had picked up the phone to Donald Trump and asked him to put a stop to it. Anyway, frankly, I think most people don't have the time to keep up with it. and maybe also don't quite know how it relates to their day-to-day work. So I think that's where Craig, you and I can jump in, talk about the latest and talk about what that then means for fundraisers and charity professionals. speaker-0: Yeah, exactly. And, you know, the landscape changes all all the time. You know, it you know, even since Fable's been reinstated, you know, up and out and now release more models. How do you keep on top of all this? And we're gonna try and pick out the essential stuff that we think is gonna gonna help you and inform you. And and then alongside that, I think the final bit we're gonna do is just share some of the tools that that we use and find helpful and talk about what we use, why we use it and ⁓ maybe give you some inspiration to go off and Try something new, different as well, ⁓ either personally or or for for your work. speaker-1: Yeah, absolutely. And I think, you know, not just how we're using it as consultants with our clients, but having quite a lot of conversations with in-house charity staff about how they're using it. So I still have a blog called the Supporter Experience Blog. And every now and then I go and interview someone in the sector who's doing something really cool and interesting that I can release on there. And we'll obviously showcase some of those cool use cases that we're seeing around about the sector here as well. speaker-0: stuff. I look forward to hearing more of that and some of those examples. I know again we get can get quite quite excited when we get talking about it. And I I do think that a lot of the possible, you know, if you can imagine it, you can turn an idea into into something real on the same day in sometimes. And it I I feel AI has has led me to to do some of the best work of my career over the last twelve months and and really ⁓ accelerated some of the the outputs for for clients and it's shown in in some of the results as well, which is which is amazing. ⁓ so to to finish off Anna, you know, what what we got on the horizon? What are the one or two things that you think people are gonna enjoy in the in the coming weeks? ⁓ I know for example, you we we're talking to take a closer look at the ethics of of AI and and charities. speaker-1: Yes, absolutely. How do you start determining which use cases really make sense at your specific organization? Because there's no one size fits all. There's no sort of rag system to say this is really useful, this is really not useful. It really does depend on the mission of your organization, the values, the methodology. So we're going to talk a little bit more about how that can be a little bit more accessible and understandable. And I'm really excited to also just take it Take it back down to basics. Talk a little bit about what's the difference between OpenAI and Anthropic and is anyone using Grok and all these different models? What are they? What's the difference? And how they being used? I think that is a great place for us to start in episode one. speaker-0: Great, and that'll be coming up. The idea is we're gonna try and release episodes weekly. We'll also do occasional deep dives and ⁓ we'd love to hear your thoughts. You know, get in touch if you've got ideas for content, if there's anything you'd like us to explore, explain, talk about, we we'd love to hear from you. And but I thought finish off an end. We've got the ⁓ our big debate next week where I'm going head head against you to talk about some of these pros and cons of AI and You know, really excited that I'm gonna be doing some research over the weekend and ⁓ you know, I've got a few hopefully a few interesting talking points and ⁓ some interesting, you know, points of view hopefully that will be, you know, again go beyond the the generic arguments we hear time and time again, ⁓ around ⁓ AI and the the sort of pros and cons of it. And ⁓ have you been sharpening your debating skills? speaker-1: this. I absolutely have. You better be careful, Craig. I've had a practice run with my sister-in-law already. speaker-0: ⁓ I'm gonna have to I've gotta I'm gonna have to up my game. I'm gonna ⁓ but yeah, it'd be be really interesting and we'll be recording that and and sharing that as ⁓ an exclusive podcast episode as well. ⁓ we better get back to the ⁓ debate prep then, Anna. We'll be back soon with the first episode. But until then, ⁓ I'm Craig and it's been lovely talking to you today. speaker-1: I'm Anna and thanks for tuning in. See you again soon.