Fireside Chat: AI for Good with Reid Hoffman

Summary

Reid Hoffman discusses how AI is a transformative force capable of reshaping industries and society, likening it to a cognitive industrial revolution. He highlights AI's potential to serve as a co-pilot for various tasks, making intelligent technology as ubiquitous as electricity. The conversation covers the societal transition from current norms to an AI-integrated future, emphasizing inclusivity and strategic deployment of AI innovations. Hoffman also addresses potential challenges, including equity concerns and the role of government regulation, advocating for an iterative, learning-focused approach to integrate AI responsibly. AI, he argues, extends beyond automating jobs to enhancing capabilities, urging public and private sectors to embrace experimentation with AI to discover its benefits responsibly and effectively.

Highlights

• Reid Hoffman likens AI's impact to electricity, making tech omnipresent ⚡.
• AI will shift how professionals, including doctors, perform their roles 👨‍⚕️.
• The importance of inclusive AI to prevent widening inequalities 🌈.
• Use of AI tools can facilitate learning and skill improvement across sectors 📚.
• Public sector innovation in AI can lead to safer, more efficient societies 🚦.

Key Takeaways

• AI is set to revolutionize industries by acting as a co-pilot for tasks 🧩.
• Inclusion and accessibility in AI deployment are crucial for equitable growth 🌍.
• AI's impact is like a cognitive industrial revolution, spreading intelligence wide 🌐.
• Iterative learning and experimentation with AI should be encouraged 🧪.
• Public-private partnerships are key in deploying AI for societal benefits 🤝.

Overview

Reid Hoffman eloquently describes AI as a cognitive force akin to a steam engine for the mind, marking the onset of a new industrial revolution. He illustrates how AI co-pilots can streamline professional tasks from marketing to medical diagnostics, boosting efficiency and innovation across the board.

Addressing the dynamics of inclusivity, Hoffman stresses the importance of equitable AI innovations to prevent exacerbating existing inequalities. He commends current efforts for voluntary commitments from tech companies to ensure responsible AI usage, advocating for cooperation between public and private sectors.

Concluding on a positive note, Hoffman envisions AI's expansive role in future societies—from smart cities to personalized education and beyond. He encourages continuous engagement and experimentation with AI to uncover its full potential while maintaining openness to adapt and improve regulations as needed.

Chapters

• 00:00 - 00:30: Introduction and Opening Remarks The chapter begins with a warm welcome, expressing delight and gratitude for the speaker's presence in Denver to kick off the conference. The speaker is celebrated as a leader and visionary in their field, setting the stage for an engaging discussion. The chapter sets the context for exploring the speaker's high-level insights on how AI is poised to fundamentally transform industries and society at large, sparking excitement and anticipation among the audience.
• 00:30 - 02:30: Big Picture Trends in AI The chapter discusses future trends in AI, particularly focusing on the idea that every language-based task will eventually have an AI co-pilot. It emphasizes the pervasive role AI could play in tasks beyond simple writing, such as coordinating meetings and handling financial tasks.
• 02:30 - 04:30: AI as a Co-Pilot The chapter explores how artificial intelligence (AI) can serve as a co-pilot, assisting in various professional activities including legal work and medical diagnosis. It suggests that any device with a computational unit, such as phones, computers, cameras, and even cars, is expected to become smarter with the integration of AI.
• 04:30 - 06:30: Cognitive Industrial Revolution This chapter, titled 'Cognitive Industrial Revolution,' discusses the transformative potential of intelligence as a technology that could be disseminated similarly to electricity. It likens this era to a 'cognitive Industrial Revolution,' comparing the impact of artificial intelligence to that of the steam engine during the original Industrial Revolution. This comparison highlights AI's potential to significantly boost productivity and contribute to societal advancements, such as the creation of a middle class and literacy.
• 06:30 - 09:30: Benefits and Challenges of AI This chapter discusses the benefits and challenges associated with the transition to AI-driven systems. It acknowledges the complexity of the transition phase but maintains a positive outlook on the eventual outcomes. The chapter draws parallels to the Industrial Revolution, highlighting the need to manage this transition more effectively than in the past. It mentions that jobs, such as those in marketing, will evolve with AI tools assisting in decision-making processes, such as selecting marketing slogans and images.
• 09:30 - 13:00: Disruption and Breakthroughs in Various Sectors The chapter titled 'Disruption and Breakthroughs in Various Sectors' discusses the rapid advancements in technology and how it significantly impacts industries, particularly the tourism sector in Denver. It highlights the use of tools like Aon that have revolutionized the way tasks are performed. Tasks that previously required a team and a significant amount of time can now be completed rapidly by an individual, demonstrating a major shift in efficiency and productivity.
• 13:00 - 15:00: Importance of Inclusivity and Equity in AI Innovations This chapter discusses how inclusivity and equity are crucial in AI innovations. It begins by acknowledging the challenges people face in learning new skills, and how AI can assist in overcoming these challenges. The chapter uses the historical example of the creation of VisiCalc and Excel, tools that transformed accounting. Initially, there were fears that accountants would become obsolete as these tools automated calculations and ledger maintenance. However, it argues that AI, like early spreadsheets, should be seen as a hopeful lens for enhancing human capabilities rather than replacing them, especially if developed with inclusivity and equity in mind.
• 15:00 - 21:00: Guardrails for AI Development The chapter 'Guardrails for AI Development' discusses the transformation of accounting through software and computers. Traditionally, accounting involved manual processes like using ledgers to add columns. Now, the role includes strategic analysis, addressing scenarios, identifying vulnerabilities, and thinking strategically about numbers. This transition exemplifies the broader changes happening in various fields with the integration of technology.
• 21:00 - 25:00: Public Sector's Role in AI Innovation The chapter discusses the evolving role of AI technology, drawing parallels to the internet's early days. Initially perceived as limited to niche applications like helping with homework, AI is expected to become ubiquitous over time, similar to how the internet became integral to daily life. The narrative suggests we are in a transitional phase with AI, where its potential is yet to be fully realized.
• 25:00 - 29:00: Private Sector Obligations in AI Development This chapter introduces the subject of private sector obligations in the development of artificial intelligence (AI). It begins by exploring the sectors where AI is expected to bring significant disruptions or breakthroughs, generating hope and excitement. The narrative highlights the revolutionary potential of AI and how it is perceived in public discourse, particularly by the press.
• 29:00 - 33:00: Intellectual Property in the AI Age The chapter delves into the evolving landscape of power dynamics influenced by technology companies, especially in the realm of artificial intelligence (AI). It discusses the role of the press and government in regulating tech companies to prevent them from gaining excessive control. The author, who is releasing a book titled 'Super Agency,' aims to address these concerns and offers a new perspective on these challenges. Furthermore, the chapter reflects on the role of government from smartphones to AI assistants and highlights the need for regulatory oversight in the tech industry.
• 33:00 - 39:00: Early Adoption and Public Sector Engagement with AI This chapter discusses the growing integration of AI in the public sector and healthcare, highlighting how AI can provide immediate medical assistance potentially at low costs. It suggests that while AI tools may not replace doctors, they can complement medical staff by offering quick advice or help, particularly in situations where immediate access to a doctor is unavailable, whether due to being uninsured, remote, or time-sensitive circumstances. The chapter emphasizes the potential for AI to be a readily available resource that can be developed with current technology.
• 39:00 - 46:00: Impact of AI on Future Generations and Education This chapter discusses the transformative impact of AI on education and future generations. A key point is the concept of AI as an infinitely patient tutor, applicable to every subject and age group. This advancement means anyone, from children to professionals, such as marketers needing to learn about AI, can access personalized learning experiences. AI enables continuous professional development and adaptation, ensuring individuals can thrive in their fields by leveraging AI technologies. The chapter highlights AI's potential to revolutionize traditional education paradigms, promoting lifelong learning and adaptability.
• 46:00 - 53:00: AI Solutions for Traffic and Public Safety The chapter titled 'AI Solutions for Traffic and Public Safety' discusses the importance of making AI tools accessible to everyone, not just the wealthy. It highlights the democratization of technology as a crucial step in ensuring equitable educational opportunities for all, similar to how tutors have traditionally been available only to those who can afford them. The text implies a shift in thinking and action towards broader accessibility and inclusivity in AI applications, aiming to make significant enhancements in public safety and traffic management.
• 53:00 - 59:00: Experimentation and Learning in AI Adoption The chapter discusses the transformation in how users will interact with phones and computers due to advancements in AI. It highlights the shift from traditional methods of operation, where users had to learn specific languages and methods to operate devices and applications, to potentially more intuitive interfaces enabled by AI.
• 59:00 - 60:00: Conclusion and Closing Remarks This chapter explores the transformative impact of generative AI on the way we interact with computational devices. It highlights that unlike traditional devices, generative AI allows for more natural and flexible interactions. An example given is how one might ask an AI system to recommend a new and interesting restaurant in Denver, illustrating how AI can adapt and respond to user preferences in real-time, navigating beyond rigid presets. The chapter closes by emphasizing the potential of AI to enhance daily decision-making and exploration of new experiences.

Fireside Chat: AI for Good with Reid Hoffman Transcription

• 00:00 - 00:30 [Applause] re welcome to Denver thank you I love this town so I I I'm always delighted when you invite me uh delighted to have you back and so grateful to get the chance to get your thoughts and insights uh as the kick off to this uh conference we'll Jump Right In Hope people excited to hear from you um obviously you have been a leader and a Visionary uh in this space uh for a long time uh tell us at the highest level about how do you see AI fundamentally transforming Industries and society as a whole what are the big
• 00:30 - 01:00 picture Trends we should be thinking about and prepared for as we look at what's possible well it doesn't surprise you I could literally spend the next 50 minutes answering my question so I try to be a little brief um so there's a couple different ways of looking at it one is uh every task that we do with language will have a co-pilot so and we do a lot of tasks with language it's not just you know kind of writing memos and so forth uh we coordinate meetings we do Financial an
• 01:00 - 01:30 analysis we do legal work we do medical diagnosis we do all of these things everything that language uh that will be a co-pilot for for professional activity another way of looking at it is that anything that has a computational unit in it is going to get smarter and so you know you naturally think out your phone your computer but by the way the camera you know the cameras that are filming us right now speakers Car
• 01:30 - 02:00 you stop lights eventually you know Etc and and so intelligence will get spread like electricity now I know that people get because it's such a transformative technology and one of the ways I describe it is a cognitive Industrial Revolution so it's like a steam engine of the mind and you know the Industrial Revolution was enormously beneficial for creating a middle class uh for creating the kind of productivity that led to literate societies that all that the
• 02:00 - 02:30 transition is challenging and we have I think that transition I think the the end state is very very good I think we have to try to navigate the transition better than we did with the Industrial Revolution now the good news about it is you say hey my my job as a marketer is going to change because now I'm going to use this tool that when I'm thinking about like okay which marketing slogan and image should I use to way you know
• 02:30 - 03:00 Denver is amazing uh place to come for tourism industry and you're going to use Aon and you're going to get 20 different answers very quickly and you're going to be sorting through going okay uh thing that the thing that would have taken me like a team of three to seven people to do in a week I can do the first prototype in 30 minutes on my own and you go oh my God does that like totally changeing well uh it does so you the
• 03:00 - 03:30 person has to learn how to do new things and one of the benefits is AI can help you learn to do that so the how do I learn my how do I learn the new skills how do I learn AI to do this well but as a kind of a lens of Hope when um vial and Excel were created uh people thought okay accounting accountants are going to go away right because what accountants used to do is they used to like have these Ledger books and write down and do all the the addition and to make sure
• 03:30 - 04:00 the book was exactly right and soth and that task went with it's all done by software and computers however accounting just transformed right accounting is still there because now instead of saying hey what I'm doing is I'm sitting there with my little Ledger machine adding up columns and thinking okay what's the analysis of what are our scenario plans what are our points of vulnerability how should we account for some of these numbers how should we think about this strategically and I think that's the kind of transation we're going to see in
• 04:00 - 04:30 all of this work and then the tool can help us with it and that's part of the reason why I think you know um you know we're currently in a oh look you know you said this AI thing was going to be big and all it's really doing is is is helping my children with their homework and it's like yeah that's a natural thing by the way remember the internet you know first it was going to be everything and then oh gosh there's nothing here other than some websites and some you know kind of uh entertainment things and now it's everywhere and that pattern I think we are in with AI Shar what are some of the
• 04:30 - 05:00 sectors that you're most excited about there being disruption or kind of Step function breakthroughs that you think AI are bringing what are some of the phic examples that come to mind that give you a lot of Hope and excitement well I think um I'll start with a different kind of a an angle into that question which is if you listen to a lot of the press this course it's like well what's really important is this AI thing is going to be revolutionary is going to create uh
• 05:00 - 05:30 an alterate landscape of power and so the primary thing the press and government should be doing is constraining uh Tech company's ability to do that and I understand the concerns and actually I'm publishing a book called Super agency uh to try to address those concerns and give a different lens of thinking about it but um if if I thought what is the actual role of government in this stuff is we have line of sight on smartphones to AIC assistant
• 05:30 - 06:00 on every smartphone that can run for you know call it a couple of dollars an hour that doesn't mean doctors going be out of work because by the way the Medical Assistant Plus the doctor be better but there's all kinds of times you have access the doctors maybe you're uninsured maybe you're you know you you fell down you're hiking in one of the beautiful hikes around here and need immediate access maybe it's 11:00 p.m. and you're worried about what's happening with your kid right there's all of this thing that's line of sight we can build it today right so that's
• 06:00 - 06:30 super exciting and then when it gets to education um part of the kind of the education thing is you have an infinitely patient tutor for every subject every age um and that means that you know that the same thing of like okay I need to learn now I'm a marketer and I need to learn now how to use AI to be a to to to to continue to be a professional in my job well tutors to do that so those are instances that I think
• 06:30 - 07:00 we should want as soon as possible right and and the important thing is to make it available to everybody right um uh you know people have been able to you know wealthy people have been able to afford tutors for you know since you know maybe all time um and you know different definitions and forth but now it's like well for everybody and and that's you know that that's the kind of thing that I think we should be wanting now to go to just kind of provoke your
• 07:00 - 07:30 mind a little bit part of what's going to happen is your interface with a phone and interface with a computer is going to completely change because right now what we've all learned every everyone in this room including myself is we've learned like that very specific language about like oh this is how you open the folder this is how you launch the document right this is how you use you know word or something else in order to do that and and we've all been learning that we've all been constrained in how
• 07:30 - 08:00 we use computational devices because they require very rigid semantics in order to function well now with generative AI you're going to be able to do something like go to your computer and go you know um what would be a really new interesting restaurant to go to in Denver tonight like I I love my old standbys but like is there something new and interesting what would you recommend to me and he we go figure it out and present it to you and that's just like
• 08:00 - 08:30 an instance of you're just talking to it like you would talk to a human being and it can generate all of the necessary work to make that happen and then if you wrap that back to for example like a medical assistant you go huh you know my lab is vomiting right um which should I be concerned what should I do know etc etc for for making that happen and that's there and it's AA it's available
• 08:30 - 09:00 to you right then and by the way part of having a good assistant is it might go oh that you should head to the vet right now and that's useful to know right good side emergency yes the um you mentioned a little bit about the potential uh Equity implications on things like tutoring I think one of the uh questions people may raise is around how we make sure that the AI Innovations are inclusive and Equitable that there's there a risk of expanding inequalities rather than narrowing them uh how should we think about and what guard rails we
• 09:00 - 09:30 want in place or what incentives in place to make sure that the technology uh is driving inclusivity and Equity as opposed to exacerbated well this is one of the things I really like about how the Press discourse Works um as opposed to the let's slow down and let's try to enshrine the Past against the future it's the let make sure the future is inclusive and here are some concerns about how to be attentive and inclusive in the future and I think that's a good thing and I think you know broadly um
• 09:30 - 10:00 one of the things that people don't sufficiently usually appreciate about businesses is businesses tend to be pretty responsive to customers and pretty responsive to communities kind of yelling at them um and and so that's a good thing and then they'll they'll be figuring out what do we do for that and as instances that this is already happening if you look at for example you know we we we're introduced by uh you know the The Honorable gentleman from Google look at Google Google provides these translate features
• 10:00 - 10:30 entirely as a kind of a pro bono project it's part of a demonstration I don't make anything from it and part of that is is we should be giving guidance to Google Microsoft Amazon all these companies say this is the kind of thing that we would really like to be helping our society Because by the way they'd like to be the good guys right and you know part of that uh you know part of that is say well what are the things that are important to be inclusive and you're not going to be able to be
• 10:30 - 11:00 inclusive in everything and you're not going to be able to be inclusive in everything the day one it's an iterative Dynamic process by the way iterative Dynamic process is exactly how we work in all technology I mean if if um you know the car had been pitched as a two-ton death machine that can run over children and possibly you know kind of cause you know major traffic accidents and everything else we'd never have cars and yet if you think about the empowerment an agency and modern society
• 11:00 - 11:30 cars are essential you know just about in almost every certainly every major country almost every country and so you have to have an iterative thing and that doesn't mean you you won't figure out seat belts until you're you figure out you won't figure out airbags until you're in it but but it's a dynamic process as you're going and that's why it's an ongoing dialogue so the notion of inclusion is not we're going to stop everything until everything's perfect minute one is we're going to be doing itative deployment and then we're going
• 11:30 - 12:00 to be going hey this is really important to change this is really important to to do stuff and I think that the most important thing is that when you think about how the new set of skills are going to be how to use AI to do all this stuff let's make sure that you know the you know one of the things that we love about you know kind of the American dream the broad-based you know talent in all communities has access to learning and potentially do it and potentially have career paths and jobs and all the rest around I think that's the at Le at
• 12:00 - 12:30 least that's one of the areas that I'm most focus on uh let's lean into obviously you are an optimist uh uh about the opportunities what AI can bring I know you spent some of your life addressing folks that are pessimists or that are concerned about the risks what are the right set of guard rails or protections that you think we should be mindful about deploying as a government if if you think there are and how do you address the concerns that folks have around the potential downside risks well I thought you know Comm Harris with the executive order around AI did it nearly
• 12:30 - 13:00 perfectly and part of that doing it nearly perfectly was I mean Biden you know doesn't really understand AI that well so hand it over to kamla kamla said all right uh I'll drive this and um and so you know brought a whole bunch of the the leading companies and said what are the things that are the most important issues what are the things you will voluntarily agree to do in order to address risks challenges Etc and almost like water marking of AI generated
• 13:00 - 13:30 content or misinformation and a bunch of other things that kind of went into that then they looked at that after having put some you know kind of I think everyone in this room has done this or participated in this you sit down with you know when you're sitting down some you're saying you can do better you can do more you can do more that was the what the voluntary commitments you know the they're pushing them to get as as strong as they could then they looked and said okay so what do we have the legal Powers going to do that are most important and part of what they resolved to which I think is the really smart thing was was they said look there's a
• 13:30 - 14:00 couple of really big risks that we need to navigate and everything else we should be learning so we want you to say tell us you have a safety plan tell us you're doing red teaming and then give us information about what's happening what's doing now um I would have recommended this to the Colorado legislator legislature because you know part of you know recently signed bill is let's try to prevent the future of software from being in Colorado which doesn't strike me as particularly smart
• 14:00 - 14:30 play uh I think that the important thing in it is is to say learn like don't try to say I have big stick with liability because by the way the tool set to do more important things for businesses for Citizens for workers for society the tool set's going to be getting richer you want that tool set to be developed and then you want to be going well this one needs to be more inclusive and this one needs to be tuned out a little bit but that's the process so it's a dialogue of learning as opposed to the I
• 14:30 - 15:00 have an Amorphis um uh poorly defined big accountability stick which will quell investment and deployment turns out to be not the strategy um and I think it's actually credit to the uh community of Business Leaders here that rallied I think after that to work on some changes and improvements to make sure that uh that gets addressed we're very excited that Denver will be very much open for business in this space as we go Colorado um uh let me ask you uh a question I mean you you DES use the phrase homot
• 15:00 - 15:30 sometimes idea that we are essentially tool Builders uh as a species one of this is about how we learned to use those tools and so if you were talking that we have public sector leaders throughout uh the audience here today people who run Departments of Transportation infrastructure who are School superintendents who are police Chiefs who are Mayors and chief operating officers of cities how should public sector leaders be thinking about how on a daily practice we should be using and deploying uh AI as part of our
• 15:30 - 16:00 problem solving so the technology is being built in a lot of different ways and so um I'll start with the advice I give individuals and then I will go to the kind of the public sector the advice to give individuals is start playing with it and start seeing on things that matter to you like work that matters to you not like oh look it's great to create a Sonic for your kids' birthday do that but but like what's what's
• 16:00 - 16:30 something that matters to you in the work you're doing and you'll find some of it still not there like for example with my investing hat you know when I use you know gpd1 or gp4 it doesn't really help me you know it helps me with a due diligence list for technology areas that I understand that well but like for example one of the things I did very early is I said how should Reed Hoffman make money investing in AI as a question to T before um I wanted to see and it basically gave me a business business school professors answer where
• 16:30 - 17:00 the business school Professor is very smart but doesn't understand Venture Capital right so it's like okay that one doesn't work Y and you'll discover those as you do those but what you but but like Ethan mullik uh of warden has a great line which is the worst AI you'll ever use in your life is the AI you're using today it's being developed so so being engaged in it learning and trying things that kind of matter to you um and that's part of the reason I was starting with like like even today like if you have some concern with your
• 17:00 - 17:30 dog or cat you can go to these these AIS and ask it and you can get something it's actually kind of useful right so you know try it on things that matter to you and that includes your work and all the rest now when it gets to public sector stuff there's there's a couple different I think principles that are important um one is to have the public sector kind of directing you know incenting enabling the tech development to be very good for the city the the county the state the country um and so
• 17:30 - 18:00 that's part of what I was saying earlier is like like going to the tech compan say well it would be really helpful if you did this kind of stuff for us U Because by the way the companies want to be the heroes and they do by the way want you know an ability to navigate uh available customers you know kind of um kind of questions of development and liability they wanted data like all this stuff and they want to be on the right side of it so I was like hey we can do that if what you're doing is helping us with problems that we're trying to solve
• 18:00 - 18:30 for our community for our citizens and I think that basically breaks into kind of call it like more boring things and more kind of Sizzle things and they're both important but the boring things are like well you know we have a bunch of different programs I mean you know this a million times better than I do but there's a bunch of different programs to help um families children you know communities you know whether it's you know food education uh public transport uh Public Safety all
• 18:30 - 19:00 of this stuff well all of that can be made like much easier with AI because for example like one of the things that I see a whole bunch of people doing on like the medical thing is you know something like 60% of doctor's times is filling out all the forms blood post phit well yeah I can do that today right and you get too extra productivity from your doctor your doctor can SP twice as much time with you right it's a great thing so like do that and that's the
• 19:00 - 19:30 same thing in the public service and that's what I mean by the kind of the boring things of you know the the the the the the awareness of of these Services the ability to engage with them the ability to to use them and then I think there's some really interesting things in terms of Sizzle which is you know one of the things that I think is only a question of time is when Society say we want the equivalent of the public service app pre-installed on the the home screen of every phone because we
• 19:30 - 20:00 want to be able to do it with hey you see a Poole take a picture send it to us let us know let's have the network of sensors and people help it hey you see something happening like a fight or something else make an alert right you you you can be a version of public private partnership with the whole network of of citizens right for how how it works and all of that kind of stuff can be extremely important for a well-run city a well-run community um you know people sense of safety
• 20:00 - 20:30 people sense of opportunity and so I think that I think it's literally only a question way now obviously I said app but part of it is you have a um and by the way I don't think it'll just be one agent on phone I think you have multiple but you have agents on the phone and one of the agents will be is hey I'm encountering this problem I I have a flat tire uh how can I get help you know and it's like okay that's useful do as opposed to you know today it's like well I I have something in the trunk and I don't have in the trunk I'm going to be
• 20:30 - 21:00 kind of trying to wave at all the cars that are passing by and hope that a good samarit will stop PR also you know like start using my directory of my phone and kind of go okay is there tow truck and call you know anyway the um uh maybe funny visual but I think in the world as we saw sort of previous to this is uh the public sector and the I sector were sort of like two folks awkwardly at a party across the room not willing to get talk to each other and the question is how do we help make that match I want to ask you a question a little bit from both sides like what does the public
• 21:00 - 21:30 sector need to do to make it easy and attractive for AI innovators companies um to be inspired to want to build products and work with us what what is Our obligation uh to walk across the room first and I'll ask you the private sector version minute well as I said earlier the key thing in this stuff is the dynamic and learning so like when you called me and said hey I'm going to have this city AI Summit I was like I absolutely want help do that because I
• 21:30 - 22:00 think it's part of how like in my book impromptu I described AI as amplification intelligence it's kind of human amplification I think that's exactly the kind of lens we should be doing and so these dialogues are critical and then at the next level of specificity I think that you know part of what you have to Prior is there's a ton of things these companies can be doing with their teams and their compute budget and companies tend to be most motivated by which of the most valuable
• 22:00 - 22:30 franchises EG with high Revenue High operating margins that they'll have over time and they're playing a theory of a strategy a theory of the game and how to get there there's an offense thing like new products new strategies and defensive holding on to the ones I have that's how to predict even startup companies although startup companies tend to be the metaphor I use to them as they jump off a cliff and you assemble an airplane on the way down because by default you're Deb and it's one of the reasons why entrepreneurship and and and and and startups are so important which
• 22:30 - 23:00 you know I actually saw on your uh sponsor list like Endeavor is here and Endeavor is great for that because entrepreneurship is how we create those new things um and so you know I think that the question about the public sector is say we want you to enable you to do things that fit broadly within your National incentives um and it could be you know Revenue although usually revenue is not
• 23:00 - 23:30 for the tech companies revenue is is usually much better from companies than from governments and other kinds of things so it's you know it's a little bit thing but like hey if you develop this product with us that will help you build all these other products and and we will help you like navigate getting this product to Market we will help navigate political concerns Community concerns and we can play partner with you in doing that in order do that and that's you know whether it's education
• 23:30 - 24:00 um you know how you and I got to know each other trying to figure out how to help uh Public Schools uh or um you know anything else that goes within the the the kind of the Civic responsibilities again like for example like one pitch might be hey this is a really easy application on top of stuff you have and it could be a really good Showcase of how you're helping cities would you just invest in it because it's
• 24:00 - 24:30 useful for communities to understand that your company care about cares about them and this is one we could help with and by the way sometimes companies go yeah that's not too expensive we have a team we'll do that and and then it's kind of the dialogue is what figures out which things are the the first things to do um now I'm not uh close enough to uh the problems that you know you as mayor uh navig to to have anything Pine off
• 24:30 - 25:00 but you've got some excellent people like fedra like fedra is the person I would call to ask that question we did lucky to get to see her today so she's coming up soon so uh so anyway but I think that's the that's the process and and the thinking of like people tend to think oh public private partnership is that a corruption and it's like well it can be if it's give me a whole bunch of money for delivering no services that matter to people that's like a bad outcome but it can be awesome which is
• 25:00 - 25:30 get like just like for example the internet and like search and other kinds of things give me a bunch of good things that can help my citizens and it it's a win for both of us so I think it's totally doable and I think it's exactly the right thing to focus on and you talked a little about this but the uh the other side of that how we what is the obligation of the private sector innovators in this space to be able to reach out and make it easier for public entities to partner in a time where some of their residents or constituents may have concerns what is your advice to
• 25:30 - 26:00 those folks who want to build that bridge what what are the things they can do to make it more likely public enties and partner with them well I think it's important to keep an open mind um you know one of the things that a lot of um technologists in the current ERA kind of feel a little beleaguered of you know like I'll be hyperbolic here you're creating AI that is already damaging our democracy it's going to challenge our jobs and
• 26:00 - 26:30 otherwise kind of Ruin society and you're doing it for your profit margin we'd love to talk to you and you're like that is not why we're doing this right uh and if you take a look at it just look at what products they're building and what kinds of things they're doing and how it works and you know kind of why for example from a Microsoft perspective co-pilot is such a an emphasis of how do you enable the human being to be have more agency and more power and and capabil and yes you know there's a whole bunch of programs
• 26:30 - 27:00 within Microsoft about how to be inclusive there's you know hundreds of people who are employed by Microsoft to try to solve like make sure that inclusion is part of the earliest design thinking and and what you create and that you'll get there uh iteratively over time and so um so I think that's uh so that's frequently a beleaguered state that a technologist finds themselves in and so I think the encouragement is say look um yeah find the right people who want
• 27:00 - 27:30 to create the future with you who aren't saying uh technology in the future is just worse than life now that with technology in the future we can create something that makes a big difference for people um and I want to do that that doesn't mean I'm blind the risks and don't have navigation but I want to I want that outcome and I think that it's a good thing for um private sector folks to keep their minds open to this which is exactly the reason why when you
• 27:30 - 28:00 called me and said hey we're doing this s can you come like yes so glad you did um let me ask you a question about um intellectual property in the AI world I thought it was so fascinating when I watched you do the interview with your digital twin where uh you uploaded all of the things you've ever written and then the digital re Hoffman got to interview you and vice versa um uh anyone obviously could go to gen tool and say write me a speech in the voice of Reed Hoffman um and they would be able to do that uh how do we think about the protections for intellectual property in a moment where that capacity
• 28:00 - 28:30 will be well I think we're going to evolve what we think and most what most people don't realize like they start with a simple Trope of you know I have data ownership but actually in fact if you look at even copyright copyright is selective data ownership there's a whole bunch of things that go into copyright like um like you know you write a book I'm allowed to use some of it for fair use if I buy the book you can't tell me that I can't resell it or hand or give it away to somebody else you know etc etc there's a whole stack of things where it's not that simple and I think
• 28:30 - 29:00 we're going to need to get to what's the new way of doing that and part of the reason why we design copyright we design patents and so forth is to have the right kind of incentives for people that do work and create IP that's really valuable for society and have valuable economic careers doing that we want to preserve that we want to amplify that but then also not say you know um because I you know found a you know was
• 29:00 - 29:30 lucky enough to find an old Shakespeare manuscript I now own this Shakespeare manuscript for all time you know as a as an that's an example not a um and so just didn't want to PR discovered old Shakespeare thing be cool but hasn't happened um the uh and so I think that the question is is to not like too often the discussion is litigating the past
• 29:30 - 30:00 and the question is what should the shape of the future be and part of the shape of the future is um it will it will change some of the Landscapes and part of what happens is people tend to want to say well I want my own circumstances to only be better but we have a term for this Lites right because it's like well like if if you know kind of England hadn't embraced the loom and said no it's only hand Weavers like I'm not sure you ultimately get to a middle class right and by the way there's a
• 30:00 - 30:30 challenge it was difficult for the Weavers now we have better Tools in modern time to help with Weaver equivalents but we need to be kind of getting to that future and that's the that's kind of the key thing that doesn't mean that the the future should be uh like all decided by technologist that's one of the reason why this dialogue is really important and if you listen to a lot of the technologists are like they all say we need to have Society in the loop talking to us in various ways to make kind of decisions
• 30:30 - 31:00 about which parts of kind of the human rights The Human Condition you know uh enablement is actually in fact really important to double down on preserve build a new you know in the next generation of Technology yeah the um let me ask a question about uh early adoption obviously one of the things we want to do in Denver is to be able to say we want this city to be kind of an open API to innovators that are coming up with big Solutions and then deploying them there is a different burden when you are
• 31:00 - 31:30 deploying solutions to public constituents and in public spaces higher risks and failure Etc how should um public entities think about uh what it means to take a minimum viable product and be able to go out with it even though it might be faulty uh and how do we communicate with the public to make them prepared for that uh so that we can maximize early adoption of exciting Concepts it's a great question and hard to solve as you know which is the reason why you asked it uh roughly speaking I think uh political
• 31:30 - 32:00 leaders can't take very much risk that can end up with egg on their face because then it becomes the um the challenge in their next election and and so for example you know federal government investing directly in solar energy companies ended up getting a bunch of people who were doing that kind of under attack even though like solar di or was wasn't like a there were private investors to wasn't a terrible
• 32:00 - 32:30 idea but those risks whenever they play not out what happens is someone Grand stands that you knew that risk was not going to work and you you failed at that risk and unfortunately we as human beings tend to have this it was either Manifest Destiny when you win or it was Manifest Destiny if you lose and it isn't like a luck probability risk taking Etc which is why we have a venture capital industry and do stuff and why we have companies because companies were designed as technology to enable The Taking of risk
• 32:30 - 33:00 so I think the part of the public private partnership thing here is to say look the company needs to take some of the risks EG if it blows up in your face the stick hits the company right the you know the it's like the you sucked you you did this badly Etc and I think that's a n by Nature how the public private has to work but companies should just be compensated for
• 33:00 - 33:30 that like for example when you look at the Venture Capital industry you know I get grief when I invest in a series a company that grows and blows up and you know people like oh that was a terrible investment that was you know one more of those Silicon Valley venture capitalists hyping their their dumbass technology and d da um by the way I I do it because I think it work doesn't always work uh but when I have other ones that work
• 33:30 - 34:00 then the economic is worth the entire portfolio and that's the reason why the company the companies are set up for doing the risk management so in the public private side I think it's the the more of the risk or by companies but you know the government can help them navigate the the public can help navigate those risks some um but then you just have to say well just like the reason why we have uh we invented companies this limited liability Cor ations as technology in pass to enable risk-taking Behavior to
• 34:00 - 34:30 to enable new Industries and new products and new Services we do that and then it just has to be an economic reward that that risk is worth it you uh talked in your speech of bolog about just to shift a minute um about Thea this will be a new generation of of uh geni natives this new generation folks will be native to AI that's different from how people might have been native to other Generations in Tech uh how will will that change the way that they learn and how will that change
• 34:30 - 35:00 the way that they enter and contribute to the workforce and how do we think about how we're changing the kind of skills we're looking for uh in people that contribute given what they will grow up be able to know and do so let me start with a somewhat historical perspective most people here don't realize that when the printing press was invented the the public discourse was this is going to end Society um it's going to ca the spread of misinformation um the experts who
• 35:00 - 35:30 otherwise are trained deeply in being how to use the information are no longer in control of it and that information is now out above about everyone and that was the discourse so not surprisingly as we go through waves of Technology both before then and after then that is the first thing that the critics were trying to help you know kind of uh navigate our human future start with now that being said think about like for example what the internet meant for how cognition
• 35:30 - 36:00 works so prior to the internet uh and by the way even more intensely in earlier times pre-writing pre-printing press memory was considered to be an extremely important part of intelligence because well there is no access there's no Library there is no access Etc so so person one is a lot smarter than person two when person one's memory works a lot better and so that was a really important part of intelligence now you come to today and you go well how
• 36:00 - 36:30 important it is to remember what was on page 34 of that book that's not fairy you just have to remember that it was in the book and then you go find it right and so intelligence changes and education therefore changes and so part of the the speech I was giving at bolognia um last year was to say we want to think about this is how we're enabling um our you know next generation gen to uh accomplish the most kind of
• 36:30 - 37:00 humanist of tasks including learning because that uh questions around like how does my intelligence work what's most important and so forth I I'll give another parallel which is when gbd4 launched you know U basically you know what you can hear from most of the US was the human Cry of oh my God everyone's going to be cheating on their essays um and I was like well look if I was a teacher and I said I'm going to
• 37:00 - 37:30 teach a I'm now varing in area that you have a lot more expertise than I but what what the hell right um if I'm a teacher and I'm saying hey write an essay on Jane Austin and Reflections on gender and colonialism what I would just done is gone to gbd4 created 10 essays from that and say and these essays are D minuses so you got to be better than this right so go use the tool the but you got to be better than this right
• 37:30 - 38:00 change the curve yeah and so cuz then by the way they have a tool by which they can start doing it they can start understanding what does a really good essay look like when what is a good Insight count what is a what is like when you're doing the well on the one hand you have argument a on one hand you have argument B and here's how you resolve the tension focusing on that versus the mechanics of compare and contrast essays or other kinds of things and I think that's part of the technology of the future and that's part of what we'll be having with you know that's one of the reasons why they younger folks are always much more uh
• 38:00 - 38:30 native to the new technologies digital natives internet Natives and now ai natives than the older folks yeah the um uh one of the things I love is listening to you think through U ways that this can be disruptive Innovative transformative for sectors people haven't thought of before so I'm going to just um give you a couple of just random uh intractable City problems that we deal with all the time for you to respond to I'm like he did not prep me PR you for this at all I know you're always ready for it um think about like
• 38:30 - 39:00 what what Innovations do you see possible in this space or how should be thinking about how AI can be deployed on this space you mentioned one earlier let's just take that old attitude traffic like what is possible or how should be thinking about what AI can do to help us all the problem is common is traffic well um there's a ton of stuff so one of the portfolio companies that I've invested in is called NATO and one of the things they have learned is 10% of
• 39:00 - 39:30 the drivers cause over 50% of the accidents and if you think we know who you are Yes actually one of the problems is we don't this you were getting to where technology and AI can can make a difference it's like which drivers should you make sure are the ones that tend to look at their phone when they're driving you know distracted driving is an issue it's one of the things that you know kind of natto focuses on um you know what kinds of things could you
• 39:30 - 40:00 reprogram the Light Grid to get out of I don't I don't know what kind of you know deadlock you know gridlock happens here but like obviously a huge problem for cities well could you actually have a kind of like okay here's how we clear through it and kind of work the thing to not have oh look the car is stuck in the intersection and therefore everyone is stuck right you know those kinds of ways all of those are software problems I mean yes you need sensors and you need algorithms and you need data and you need a bunch of other stuff but that's
• 40:00 - 40:30 all solvable um and it's one of the reasons why you know in addition to n which is kind of a human networked uh how do you how do you make each car a networked device um that helps safety and you know and you know eliminate distracted driving and all the rest of the stuff I also did a autonomous vehicles because you know we want autonomous vehicles to get here especially when it gets to you know highways and trucks and else but we want
• 40:30 - 41:00 them to get here because um if everything was AV today you know the over 40,000 deaths that we have per year in the US and that's just deaths there also you know injuries and all the rest medical thing that would probably go down by you know 75 or more perc that's huge and like I wrote an essay on this very early at Lin where I said drink and driving goes from with with autonomous
• 41:00 - 41:30 vehicles goes from an evil thing that you do risking other people's lives to something that you know you might do every day because he you're not driving you're not driving it's correct see that's why I love to ask you so you have a great answer um uh how about because I think what we're uh after is both pushing private sector innovators to think about ways in which you can tool build or innovate on solutions that would have massive markets you can come up with a tool that's going to help make traffic move 20% faster there are literally
• 41:30 - 42:00 billions of dollars being deployed every day in cities around the country to solve that problem that probably could be solved much more effectively with an AI tool um another one that both is very expensive and obviously has huge human costs uh is crime how do we think about ways in which AI can be used as a tool to help support interven on the crisis for well I think one of the things that's really I was mentioning a little bit on the app I mean think if what you had is every citizen because phones tend
• 42:00 - 42:30 to be validated I mean yes there are you know prepaid phones but phones tend to be validated this one's mics this one's readed and soth and say you have the ability to um participate in a call it a 911 app right then the entire citizenry which cares a lot about like diminishment of crime safety nonviolence Etc is enabled and it's like I can go oh I'm seeing something right I want to
• 42:30 - 43:00 report it and then obviously you know part of what happens is well does this person actually seeing something are they inventing something because they hate their neighbor you know blah blah blah blah you know but AI can help with all that doesn't mean it's perfect but it means that you can actually have much quicker response you can do recordings like so it kind of becomes a and it's like a safe thing like for example if you see you know one person uh beating up another going getting involved is like risky and
• 43:00 - 43:30 challenging like if you're small or else it's it's like it can be challenging you don't know if they have a knife or a gun you know it's like all this stuff whereas you take a picture you you go I see a problem right and that allows uh response and that's part of like it's another version of kind of public private partnership so I think it's it's kind of um you know that's just the beginning I think of things that could be empowering of the citizenry to say hey we work with our
• 43:30 - 44:00 services to make our community safer uh I agree I think that's one of the places where we see real possibilities for for win-win right there's a technology an AI technology now that um some police departments use which right now we find is the most significant impact on uh reducing uh the over incarceration of people of color and reducing the total number of crimes is actually the number of officers you have on the street because officers are actually a preventive tool yes the biggest challenge to their time right now is
• 44:00 - 44:30 sort of like your nurse example of many officers you respond to domestic violence uh situation you might spend four hours writing the report uh from that event that's four hours you're not on the street and so there's an AI tool that uses the body warn camera that takes all of the audio from that conversation you're there body War camera captures it all body War camera processes it all pre-rituximab second is it's actually a much more objective uh description of the event
• 44:30 - 45:00 than you might have got from the officer herself or himself so folks that are wored about police accountability it's actually helpful folks are worry about protecting police officers time you get both and by the way another thing of enabling is oh there have been three other domestic violence calls here with these people here's the available previous information to go in with some Readiness yeah great great example um the uh uh we just have a couple of minutes left um uh as you think about um uh guidance you would have for uh us if
• 45:00 - 45:30 you were to talk to both City and leaders here uh and to the ecosystem of innovators here what are the most important things for us to be focused on uh in the next couple of years to make this a place where we can easily and aggressively adapt solutions that will make an impact well it's a bias towards experimentation right like anybody who tells you this is exactly what it's going to look like five years from now
• 45:30 - 46:00 is either diluting themselves or duding you or both right we have this major wave coming and the best way to do it is to experiment with it and that was part of the reason I you know kind of was indicating my you know my expert opinion that the that recent Colorado AI bill was bad right because it's it's it's this question of enable experimentation and doesn't mean that there won't be problems and risks and that's a little bit of the reason why you try to shift those to companies try to navigate them
• 46:00 - 46:30 and you know can bistic you know Etc but but that bias towards let's learn something let's try to do something it's like like I was really appalled when I was reading about some schools we're Banning gbd4 in our school you're like us should be encouraging people to use it right you should be learning from it doesn't mean that there won't be some like oh how do I do my homework differently and how do I how do I do all this stuff and I you know that's that's a change of work process but like how do I do that and experimenting and trying
• 46:30 - 47:00 things is I think fundamental whether it's you know your position in government whether it's citizens whether it's it's companies and you go well we're not quite sure what the risks are here then experiment small right it's it's it's it's how we make you know Innovation is spelled r right and so that is extremely important to do so helpful I think that's this is the spirit of I think how we've tried to approach this one of the things we've done in Denver is try to uh
• 47:00 - 47:30 familiarize the public with we will take big swings and our problems like homelessness we're going to try to get a thousand people off the street and set that public Target and then deliver on it um which we successfully did but there will be times we'll take big swings and not succeed but what we found is U some people like the idea some people dislike the idea we had a lot of arguments about it a lot of public Community meetings um but in the end people's satisfaction is much higher at the city about how we're approaching it I think because what they're saying is not sure if this idea will work but they're trying it yes they're going to innovate they're going to come up with a
• 47:30 - 48:00 strategy they're going to implement it and if it doesn't work you'll learn from it yes that's the spirit we want to be able to bring uh uh ladies and gentlemen would you give a huge round of applause please going do a photo very