Transcript – E3 Living and Thriving in the AI Era

Welcome

Hello and welcome to the ICT4D Conference podcast where we are exploring the use of digital technologies and data Innovations to address global development and humanitarian challenges. My name is Sonja Ruetzel. I'm at CRS and I'm interviewing experts for you from our global in-person conference earlier this year in Ghana. We're now sharing some of the key takeaways and lessons learned from the discussions and

Workshops.

Introduction

Today it's my pleasure to speak with Jeff Fortune about generative AI. Jeff, first of all Welcome to our podcast. Thank you very much for having me today I appreciate it. 

In terms of introduction Jeff Fortune is the Senior Director of artificial intelligence and data engineering at Bixal. And we are exploring how we are living and thriving in an area of artificial intelligence. 

Would you just like to add one sentence about who is Bixal because I'm not sure all of our audience might be aware of your organization. Just as a as an introduction. thank you and thank you for giving me the opportunity to talk about Bixal they're an amazing

Organization. Bixal is a human centered design and technology company that looks for opportunities and communities where we can engage and deliver social good outcomes for our partners and the communities and the people that we're trying to serve. 

Thank you, Jeff. For me the emphasis is on human centered design which I think is so

important when we're talking about AI. So let's just dive into it my first question

Who does AI impact and what are your thoughts on this? just sort of setting the scene for us.

In my opinion right now AI is impacting everyone in the world. It's a transformative technology that is in almost every part of our life. Within a 2-year period of time Chat GPT really kicked off this excitement with generative AI and has really shown people across the world how they can get access to knowledge that they didn't have access to originally and so. that's maybe there is a lot of articles that were written in English and you needed to transform that to maybe French or Spanish and now you're able to do that pretty seamlessly with different tools like Chat GPT that can really take that

content and take the meaning and understanding and give it back to you in the language that you prefer to read. Which really opens up a lot especially when you think about the Technology field. We write a lot of our papers in English and so being able to get that technology back out across the world into languages that other people can understand to allow innovators to really help their communities is priceless.

I also think you know that it is very impactful for everyone. This technology is transformative, like I said, and it's transforming today and it's going to be transforming for many years to come there will be a lot of learning required and a lot of patience

needed as the innovators transforms.

Everyone needs to have an opportunity to put their input, you know, when they're able to address how you feel about AI out there and learn.

Everyone needs to have this time to start learning and preparing themselves for AI because knowledge is power. And I think the more that you learn about AI the less scary

it seems because you're able to start to understand what it's doing. And access to

it right now is one of my biggest worries. We have to make sure people have access to the tools but more importantly the training and using of those tools, because it also help provides resource to help ensure ability for communities to learn innovate. and these tools also encourage a more sustainable long-term outcome for the different communities that we're working with.

Thank you, Jeff particular with your focus on access there's generally, obviously a concern that generative AI will further widen the digital divide, so that's certainly very important point to make.

You also particular focused on the language models and particular the language aspect so going down that route… How do you think large language models are going to impact the internet?

wonderful question and the impact on internet is we're seeing it today as well, you know, with different large super applications like Chat GPT, Anthropic, Google Gemini, Microsoft co-pilots. These applications are showing us the power of having these really clever AI systems that can complete numerous task for us like coding software summarizing your daily news feed taking really large documents and summarizing those down to an easy understandable format. And additionally the internet is going to be the most affordable way for innovators to provide AI Services across the world

world too. In the near future it will cost significantly less to host AI for inference. and what inference is, it's the process in which generative AI Model, a large language model, provides you a response based on the task you give it.

Nvidia right now has really been helping build out a lot of data centers across the world is capacity with their large and more powerful computes and how this really helps is it can get down to the mission level now and the missions are able to provide Cutting

Edge AI solutions to the communities who they are serving. this is important because some of these communities still lack stable Cloud infrastructure a good internet connection having ownership over modern devices like iPhones and

different Androids and even potentially having the correct compute power within

the Regions they're currently in.

AI has to work for everyone so we need to really ensure that technology that we're

building works in these communities and that it can work for everyone that we're

serving well.

thank you Jeff and I would say now let's talk about some actual applications and make it very practical. Do you have some examples for us?

 AI has been a part of building very helpful tools and automation for a very long time. Mass production was created to reduce the cost of goods autonomous driving to

improve safety and the smartphones we carry around our hands to help with communicating. and I think building applications is great but sometimes we need to understand how to build those applications. what I would really like to focus on is the access part of this I'd really like to talk about an organization called Hugging Face that

is in a machine learning community and they provide free access to AI tools and

information and learning opportunities. no matter your current walk maturity

with your AI practices. so you can know nothing about ai go to this community

and you can really start understanding what is a prompt what is a large language model how do you generate that and they also have some really clever tools that a really large community builds so you can also see AI at work and you can test out a whole different bunch of projects.

this this open source organization hugging face provides a free service also that has a world class chat application that's much like Chat GPT they allow different several open

source models so you can go in and you can discover which model is maybe best

for your task and they also allow you to create personalized AI assistance to help automate routine tasks that you might do like writing emails summarizing

documents. which can really help you get your hands on Cutting Edge technology for free in a community that is willing to Foster help and train you along your AI journey. and also this community really helps with the fairness of innovation that has the potential to improve Services throughout all the communities around the world by ensuring the human part of AI is being represented. Hugging Face provides access

and knowledge freely which we need to ensure diversity is being represented in

our AI solutions that are being built.

thank you Jeff I think it sounds exciting and I love the fact that we have more solutions available now and I particularly like your focus on on learning about Ai and fully

understanding it. I love that that's really important just as we said in in terms of making it inclusive and I also really like the focus of hugging face on making it available and your focus on access.

 next point was just to see if we had any closing remarks. I don't know

Jeff if you had anything you wanted to share as a closing?

 I would say AI is a journey and the first thing I think you really need to focus on when you are thinking about building AI is the data that you have understand the data, understand the consequences of using that data if there's a lot of biases in that data personal information that's in that data. you want to ensure that you're protecting the data and you're protecting the people that you're working with and their rights thank you

I love that we're closing with AI ethics and AI safety as that is really important and I think another major concern or major issue we need to talk about just as well as the digital divide and AI. so thank you so much for touching upon so many important aspects in our first conversation about gen AI and thank you very much for joining us and for sharing your insights. we're adding the link to hugging face uh in the description for anybody who would like to have a look and also if you like to learn more about Jeff and about Bixal's work there will also be a link. awesome

thank you thank you very much byebye. for our audience we have more information and more events coming up uh talking about General of AI so this is a wonderful start. I hope you join us again next podcast or other channel of the ICT4D Conference.  thank you.