Given the important nature of the social ramifications of the tech we're creating, this post is going to cover the social/political aspects rather than focus on the tech/entrepreneurial aspects of the talk.
The near future is quite concerning to everyone. Causes, effects and possible solutions are discussed: the wage and prosperity disparity as seen as a factor in the past election, humanitarian consequences of AI automation and pre-requisits to build smarter cities.
There isn't much research on Universal Basic Income (UBI), or what happens after jobs are eliminated due to better technology automation despite broad support for UBI. Nellie commented that people think it's strange/inappropriate for a startup investors to do an experiment on families in Oakland. He thinks it's strange not to want to study what happens. Sam: 'We're doing this thing it's working really well, it's going to cause massive effect on everybody and how can we responsible citizens of the world.' 'We feel we have an obligation to contribute to the policy discussions with some actual data.' In an effort to gather data, YC Research is conducting a research project on UBI with an initial sample size of 100 families to be expanded to 1000 in Oakland. Sam: 'It comes down to wanting to know what happens when you take away the need for absolute survival, and what happens in reaction to the need or drive to struggle to survive, how will people live and spend their time?' In creating the research project, YC encountered pushback from some local governments of prospective cities. It's critical that the local government of the city that housed the research participants is fully supportive of the study. In order to not change the study, he's not able to elaborate on the current progress of the study. They will share the data after the study is concluded. Ultimately the they think the government will be leading these types of studies in the future.
Sam: 'This is a mega change of society, we have evolved to struggle to survive basically for a long time, in different ways and if you take that away from people and if you take that need or drive or whatever that is, what's going to happen? Again, I think it will be great, I'm optimistic about it. But it is a big change, would be good to have some data, on how it impacted people before we're in a situation where we have huge amounts of jobs limited in a small number of years, and have to make an on the fly decision.'
Nellie (and Shannon via twitter): 'What skills are AI proof?'
Sam: 'In the short term, all repetitive work that doesn’t require a human emotional connection is likely going to be greatly supplemented by machine learning. There will be some jobs that need a human connection [like human learning, hospital care]. In the future the definition of repetitive work will move down the field.' Unfortunately he thinks we'll find much of what we do is quite repetitive.
Nellie: Do you think this to partly to blame for populism?
In short Sam agrees, and surprised at how people blame both globalization and technical automation together. He said none of the nationalistic methods work against automation.
Sam: 'Struck how I didn't know how to find 100 Trump voters. That alone seemed deeply problematic.' 'If I get data that my model of the world is wrong, I then go and update my model.'
On the informal organization which is still forming, on tech values:
Sam: 'Can we come up with a list of values that we want tech companies to agree too. Are there a set of statements that we can ask a tech company to commit to: won't turnover user data without due process, will fight unconstitutional or illegal orders for user data, will protect it's immigrant employees, provide money and assistance to people that are running into problems with immigration raids or the borders, committing to creating jobs not just in San Francisco but through the country, committing to pay equity, are there a number of issues relevant to the current political climate that we can get a bunch of tech companies to agree to?' It's an informal group of about 100 people. Sam: 'A clear statement of values of things from tech companies about things that are at risk from current policy actions is good but not enough, not sufficient.' It's unclear what the correct org should be named. 'Leverage is for employees, where they would say I won't work at a company that doesn't commit to these values.' However, he counters, 'I think this is a small thing and we need to do big things.'
Nellie: Do you think this to partly to blame for populism?
In short Sam agrees, and surprised at how people blame both globalization and technical automation together. He said none of the nationalistic methods work against automation.
Sam: 'Struck how I didn't know how to find 100 Trump voters. That alone seemed deeply problematic.' 'If I get data that my model of the world is wrong, I then go and update my model.'
Sam: 'Can we come up with a list of values that we want tech companies to agree too. Are there a set of statements that we can ask a tech company to commit to: won't turnover user data without due process, will fight unconstitutional or illegal orders for user data, will protect it's immigrant employees, provide money and assistance to people that are running into problems with immigration raids or the borders, committing to creating jobs not just in San Francisco but through the country, committing to pay equity, are there a number of issues relevant to the current political climate that we can get a bunch of tech companies to agree to?' It's an informal group of about 100 people. Sam: 'A clear statement of values of things from tech companies about things that are at risk from current policy actions is good but not enough, not sufficient.' It's unclear what the correct org should be named. 'Leverage is for employees, where they would say I won't work at a company that doesn't commit to these values.' However, he counters, 'I think this is a small thing and we need to do big things.'
The sentiment that Trump was able to capture is that, many Americans have been left behind. Sam feels the tech industry has some responsibility. 'San Francisco is an incredibly optimistic place in ways other parts of the country aren't.' 'We need an economy that has a lot more winners. There's a graph that has helped me understand it's a graph that shows real productive gains, and real wage growth or ave hourly wage.' Says real wage growth diverges at about 1973, and has flatlined ever since. '...Is seen as obvious proof things are clearly broken.' 'The Trump voters believe this sense that the system is completely rigged, and that normal people don't get ahead in life'. 'The idea that we need to figure out something to do, for our citizens, who's life is not going, like this idea of the american dream seems comical to, that seems really right.' He agrees with the diagnosis that the system is rigged, doesn't agree that Trump has correct answers to fix it. Disagrees with the Tech community ostracizing of Peter Thiel due to Thiels political support for Trump. Been trying to convince people to run for governor of California in hopes of having a progressive counterweight to the current administration. 'Would love a progressive version of a Koch brothers.' In passing the olive branch, he's hopeful that the administration would be successful at making gains for the average American. He's supportive of Elon Musk on Trump's counsel, in hopes that overall they will have better tech guidance. His comment to non-Trump supporters, 'Let's at least do something proactive and not just complain ...'
YC is working on afordable housing:
Sam: 'Is there a way to build houses for 100K? It used to happen, [...] that seems comical now.' 'Can we make housing super affordable? In a place that are near their jobs, and live in a city.' He thinks it's unbelievable the number of people in SF that are pro environment that are anti-housing in cities. 'if you want to have people stop emitting carbon, let them live near their work, don't make them drive an hour or half.'
Nellie: How would you to make a city built by tech folks that have a civic soul? (Shannon:) For smart cities, how do we ensure we don't accidentally create a smart divide?
Sam responds assuming we can figure out the technical questions, we want to engage with the existing residents of that city. 'We’d only want to do this in a city where we had civic engagement.' He thinks tech has done a poor job, and also seen the cities do an equally poor job. Believes there is work todo on both sides.
Sam responds assuming we can figure out the technical questions, we want to engage with the existing residents of that city. 'We’d only want to do this in a city where we had civic engagement.' He thinks tech has done a poor job, and also seen the cities do an equally poor job. Believes there is work todo on both sides.
On divisiveness:
Sam: 'Any time there is a two party system of any sort, it becomes super tribal. 50% of people would now object to their children marring someone of the opposite party. The same number, incidentally was of marring someone of a different race 50 years ago. Now that number is below 20 percent... This super tribalistic thing that supports fringe behavior is bad. There's a version of this, tech industry and cities. Rather than 80% of the people would like some sort of middle ground solution.'
Q&A:
Question: What's SV view towards ancient cultural values?
Sam: There are books that list current values, funny how AI takes on many of culture values. ... Most all back to theses needs that cross all values.
Question: Direct democracy?
Sam responds on replacing repersentive democracy, interested in the experiment but with caution. Sam: 'There are times where the crowd is just wrong, it wasn't that long ago when women couldn't vote, that interracial marriage wasn't allowed. [ ... ] What happens when people are just wrong?'
Question: How can you ensure the solution is immune to corporate greed?
Sam: 'Gets back to we the people. There are many people, we collectively have power, we get to chose our leaders, if we don't engage if we all say that's someone else's problem we won't get the future we deserve. If we say the system isn't fair, we can get outsider leaders and that's how things will change. We need to be engaged, if the principle of democracy, if regular people engaging continues.'
Question: Tell us more about OpenAI.
Sam: OpenAI, to build with no obligation to share holders, and the benefit is to maximize to humanity. [...] We worry that AI could go badly. Wanted an org dedicated to thinking about [...] AI safety.
Question: What are problems that startups can tackle?
Sam: 'Fusion, Cancer, housing, autonomous cars, [...] need more things that startups to take on. Feels it's easier to start a hard company than an easy company. If you do something that is really important to the world, many people want to help you. I often think it's easier, and and people don't try enough. I think startups are going to solve, and I hope a lot of the problems we face.' He thinks it's possible to develop Education startup, to have an big impact.
Question: Startups focused on middle eastern market?
Sam: There are a few, saw an uptick from other companies, after a successful company returned to Egypt.
Question: What about diversity?
Sam: 'We like non-track founders. This next part is more bad than good, we have 5 top women partners, one black CEO of our core program, I'm gay and I run the whole thing. I think if you don't get a really diverse group of founders, you'll miss problems that a lot of the world faces.' Feels this provides a significant edge over traditional VC firms in the valley, but would like to be more diverse. Additionally he spends huge amounts of time outside of the valley to speak to non-tradition founders.
Question: What's your advice to find not just the diversity of people, instead to address the problems that really count to the most people?
Sam: 'It's easy to get caught up in a reference problem here in the bay area. [...] Go visit slums in Africa to ask what kind of technology they want invented. Anything you can do to help something is valuable. [...] Ask what can I do that will really impact million or billion of lives, who are a lot less fortunate.' Agrees many problems to be addressed, thinks it's important to balance that view the positive change that has happened: decrease in global poverty rate in the last 40 years. 'I think we can do a lot more. A company in YC is a model for universal health care, starting for a small village in Africa.' They are working on something quick. Believe's '[it's] a great way to think about the world, and start with it there. I think about that's a great way to think about the world and having an impact. [...] They are starting with a small group for one thousand people but thinking about how it will scale to a billion people.'
Nellie: What is your 60 second idea to change the world?
Sam: 'I think how can we make every person in the world, ... Thinking about how we use technology to make every person in the world, happy, and healthy and rich enough and fulfilled to be able to do whatever they want in their life lives, and to be able to live their lives as the way they want and that is energy and AI, and biotechnology companies.'
Sam: 'The way I think about this, I can use this lever of enabling entrepreneurs, how can I enable, inspire people to use these technologies to solve these problems.'
I'm very grateful that Sam Altman and Nellie Bowls chose to speak at InforumSF by the Commonwealth Club. The more we have open dialogue the better.
