Open||Source||Data

Episode Quotes

Vinay Kumar
"I always believe in this: you don't need to solve a very large problem. Maybe it will take a lot of time to do that. A lot of resources to do that but something small, which you can have an opportunity to solve that could be very big or a fundamental for quite a bit is fantastic. Think of a scenario where your small fundamental idea is a base for another small fundamental idea for someone else."

Charna Parkey

We also want to ground it a little bit in impact we've been seeing. And I think in the financial, banking, insurance industries it's not, I would say, an even distribution of advancement. Different countries have different regulations and different appetites for risk."


Timestamps
- [00:00:00] Introduction by Charna Parkey.

- [00:01:57] Vinay Kumar begins talking about his journey.

- [00:05:27] Discussion on building a search engine for STEM researchers.

- [00:07:06] Challenges with early deep learning.

- [00:09:55] Conversation shifts to ML observability.

- [00:17:06] Discussion on simplifying verticalized AI.

- [00:22:30] Impact of large language models (LLMs) on AI.

- [00:30:58] Comparison of autonomous cars with AI regulation.

- [00:37:58] Vinay mentions his science fiction novels.

- [00:42:19] Conversation summary with Producer Leo Godoy.

“We're really trying to show that we could co-create experiences with AI technology that augmented our experience rather than served as something to replace us in creative act”.

“For every project like [LuminAI], there's a thousand companies out there just trying to do their best to get our money... That's an uncomfortable place to be in for someone who has worked in AI for decades”.

“I had no idea what was going to happen kind of in the future. When we started EarSketch... we were advised by a couple of colleagues to not do it. And here we are, having engaged over a million and a half learners globally”.

"I remember the first robot that I built. It was part of the first robotic systems... and watching these machines work with each other was just crazy."

“If you're building a product and your goal is to engage underrepresented groups, it is on you to make sure that you're educating the folks in a way that you're trying to reach.”

(01:11) Brian Magerko's Journey into AI and Robotics 

(05:00) LuminAI and Human-Machine Collaboration in Dance

(09:00) Challenges of AI Literacy and Public Perception

(17:32) Explainable AI and Accountability 

(20:00) The Future of AI and Its Impact on Human Interaction 

(22:10) EarSketch and learning: computing as a meaningful concept 

(27:18) The need for interdisciplinary collaboration to ensure AI developments are beneficial for society as a whole.

(30:02) Brian Magerko's next reshape of the future, better understanding models of collaboration and improvisation between people and computers

(35:51) Brian Magerko's advice to researchers based on his own identity and experiences

(44:20) Projects and updates related to EarSketch and LuminAI’s improvisation model.

(46:24) Backstage with Executive Producer Leo Godoy

Timestamps

00:00:00 - 00:01:23 - Introduction

00:01:23 - 00:04:30 - Heather Domin's Journey

00:09:50 - 00:12:48 - Open Source and AI Ethics

00:12:48 - 00:15:25 - Generative AI and Governance

00:23:40 - 00:26:22 - Future of Responsible AI Practices

00:35:37 - 00:37:31 - Advice for the Audience

00:37:31 - 00:46:04 - Reflection on Risk and Hope in AI

Quotes

Heather Domin

"I think that each of us individually can scan our environment and understand, you know, where can I make an impact? What problem can I help solve? What is the next thing that I can really contribute to?"

"There are absolutely ways to automate, you know, the prompt testing and many of the routine tasks that you want to leverage automation in that way so that you can actually have the humans focus on other, other things so they can focus on the critical thinking and outside the box sort of thinking that we want the humans to be focused on."

Charna Parkey

"I think that it's a hard for people getting into it for the first time to jump to hope if they've experienced something that they should fear in the past. By that, I mean, groups that have been marginalized by other forms of technology are not going to start hopeful with this new one that is is using their data without their permission.."

"If for some reason I came to understand in a month what that meant, I should be able to go back and revoke and be like, nope, I actually don't want you to have that anymore. So I think that that would help people feel better."

Check Heather's paper: On the ROI of AI Ethics and Governance Investments 

Connect with Heather
Connect with Charna

Timestamps

1. Introduction and Background (00:00:00 - 00:01:16)

2. Ethan's Journey (00:01:16 - 00:05:12)

3. The Role of Food and Agriculture (00:05:12 - 00:06:52)

4. Investment in Regenerative Agriculture and Generative AI (00:06:52 - 00:07:44)

5. Levels of AI Impact (00:07:44 - 00:12:42)

6. HowGood's Use of AI (00:12:42 - 00:13:20)

7. Consumer Impact and Corporate Responsibility (00:13:20 - 00:15:44)

8. Future of AI in Food Systems (00:15:44 - 00:20:30)

9.  Innovative Perspectives on AI Training (00:20:30 - 00:21:10)

10. Action models in agriculture, optimizing water and soil use on a larger scale. (00:24:14 - 00:25:28)

11. Discussion on integrating human cultural geography into AI models. (00:27:37 - 00:30:00)

12. Charna and Ethan discuss procurement decisions and their impact on sustainability. (00:30:20- 00:40:15)

13. The ethical implications of AI in corporate and government decision-making. (00:42:01 - 00:54:31)

14. Leo brings up the impact of AI on consumers, discussing how AI can change purchasing decisions by highlighting product sustainability. (00:54:40 - 00:55:30)

15. Charna elaborates on using AI to understand different business models and how generational changes affect consumer choices. (00:55:47 - 00:57:32)

Quotes

Ethan Soloviev

"What if we're using ecological data? What if we're training on trees and insects and animals and whale song? What kind of questions would a gen AI trained on whale song and hummingbird language ask us?"

Charna Parkey

"If we have this great translator that is Gen AI, we already have text and language to code. We can do code generation. We can already interpret this code and tell me what it's going to do. Take that code to language. Why can't we do that with some of these other senses and these other measurements?"

Connect with Ethan
Connect with Charna

Timestamps

00:00:00 - Intro

00:02:00 - Beth’s Journey

00:19:33​ - Ontologies in AI

00:21:44 - Data Lineage and Provenance

00:32:52 - Open Source Tools

00:38:38​ - Explainable AI

00:44:58- Inspiration from Nature

Quotes

Beth Rudden: "The best thing that I could tell you that I see is that it's going to shift from more pure mathematical and statistical to much more semantic, more qualitative. Instead of quantity, we're going to have quality."

Charna Parkey: "I love that because I've been so mathematical for most of my life. I didn't have a lot of words for the feelings or expressions, right? And so I had sort of this lack of data and the Brené Brown reference you make, like I have many of her books on my shelf and I often pull, I don't even know where it is right now, but the Atlas of the Heart because I am having this feeling and I don't know what it is."

Links

Connect with Beth
Connect with Charna

Learn how Andrea Brown, CEO of Reliabl, is revolutionizing AI by ensuring diverse communities are represented in data annotation. Discover how this approach not only reduces bias but also improves algorithmic performance. Andrea shares insights from her journey as an entrepreneur and AI researcher. 

Episode timestamps

(02:22) Andrea's Career Journey and Experience with Open Source (Adobe, Macromedia, and Alteryx)

(11:59) Origins of Alteryx's AI and ML Capabilities / Challenges of Data Annotation and Bias in AI

(19:00) Data Transparency & Agency

(26:05) Ethical Data Practices

(31:00) Open Source Inclusion Algorithms

(38:20) Translating AI Governance Policies into Technical Controls

(39:00) Future Outlook for AI and ML

(42:34) Impact of Diversity Data and Inclusion in Open Source

Quotes

Andrea Brown

"If we get more of this with data transparency, if we're able to include more inputs from marginalized communities into open source data sets, into open source algorithms, then these smaller platforms that maybe can't pay for a custom algorithm can use an algorithm without having to sacrifice inclusion."

Charna Parkey

“I think if we lift every single platform up, then we'll advance all of the state of the art and I'm excited for that to happen."

Connect with Andrea
Connect with Charna

Episode timestamps

(01:47)  Asa Whillock's career journey at market-leading companies and the role of open source in each (Adobe, Macromedia, Alteryx)

(04:56) Feature Labs acquisition by Alteryx and its open source roots in democratizing machine learning capabilities

(11:00) Survey findings on enterprise board members' perspectives on AI and the need to move beyond policy creation to implementation and governance.

(27:00) Applying AI capabilities and decision-making related to AI  

(30:00) The future of AI predominance, including cost reduction, open source model advancements, and the push for demonstrating business value

(43:33) Advice for navigating AI expertise and decision-making, including continuous learning, self-awareness of decision-making models, and acknowledging knowledge limits


Quotes

Asa Whillock

"I love regulation. I think it's great. And people are like, what? Why would you say that? And the reason why I say that is because I think it puts a floor underneath all of us of what do we think good looks like?"

Charna Parkey

"I think we need to, as a community, focus on meeting them where they are if we really want the democratization that is promised. Yeah, I don't know any other way to do it."

Episode Timestamps

(02:11): Robbi Armstrong's role at KeyBank and intersection with open source and AI initiatives in the financial industry

(04:06): Compliance and regulatory trends in AI for banking

(12:10): Organizational Change Management with AI

(28:00): Responsible and Ethical AI

(37:00): Financial Literacy and AI

Quotes

Robbi Armstrong

“I truly believe that if you are an organization and you are sitting back and you're not organizing a team and you're not organizing a program and you're not learning, you're not looking at education, you're not looking at change management around Gen AI, I don't think you'll be here in two years. I really truly believe that. Because you won't be able to compete."

Charna Parkey

“I think the democratization is real and I think it's incredibly important because that step in between the domain expert and the technology is very lossy. You know, oftentimes we say, well, if only I had the data to answer your question let me give you a different answer or let me answer it completely and now we can actually put it in the hands of the experts and say, well, oh, then let's go collect that data."

Links

Connect with Robbi
Connect with Charna

Episode timestamps

(05:06): State of open source in the UK  

(07:22): Importance of open source community  

(15:19): Balancing openness and regulation in AI  

(21:19): Pace of technological development and regulation

(28:21): Reliability and discernment with AI outputs

(35:24): Universal advice

Quotes

Amanda Brock

“I think the governments that are going to win, the governments that are going to have the best regulation that promotes most innovation are going to be the ones which are able to make their regulatory environment flow in the same way as the technology evolution and innovation flows."

Charna Parkey

"I think the expectation needs to change. Part of what has happened with, you know, literal text search or keyword search and just Google and things like that, is that the average person expects what comes back to be relatively factual. That it's been referenced and, you know, backlinked, etc. That's a deterministic system. These are not. These are based upon statistical likelihoods of what word should come next."

Links

Connect with Charna

Connect with Amanda

Episode timestamps

(02:15): Tacita's unconventional career path to becoming a CTO  

(07:00): Textio's practices for building AI responsibly and ethically  

(14:00) The impact of Textio's AI on performance feedback  

(17:00) The importance of purpose-built vs generic AI models

(28:00) Balancing open source and proprietary data/models  

(42:00) Advice for the AI industry moving forward  

Quotes

Tacita Morway

“When you've got a team with different backgrounds, educational, lived experiences, identity, careers, all of those things, we have those different perspectives in the room. And we're all working off of the same expectations. We can catch each other's gaps.”

Charna Parkey

“There's an interesting conversation happening, I think, in the community right now about these purpose-built LLMs. Are they as good as generic LLMs? Sure, certainly if you're not going to apply something purpose-built to something generic or outside of its domain, it is not as good. But I think some of this shows us that unless you have something purpose-built and unless you're leveraging the data in the right way, you may just be feeding noise back into the system.”

Links

Connect with Tacita
Connect with Charna