Marc Baber suggests another roundtable on homelessness

Presenter: A former candidate for Oregon’s 4th Congressional District is proposing a community roundtable on homelessness, and this time, inviting local researchers. The nominee of the We The People Party, Marc Baber:

Marc Baber (We The People Party): Oregon’s District 4 is arguably not only the most beautiful district in the country, it’s one of the smartest, with both the Oregon State University and the University of Oregon.

Something that many people are not aware of is that Oregon’s District 4 is also home to some of the behavioral sciences institutions that make a lot of difference in terms of determining what kinds of interventions achieve the best outcomes for the least cost.

And we have both Oregon Social Learning Center (OSLC) and Oregon Research Institute (ORI) here in Eugene, Springfield. And although their work tends to focus on different areas of behavioral science than homelessness, the kinds of techniques that they’re using are the kinds of techniques that I think more people should be aware of.

And realizing that we need to rely on those when we’re talking about political solutions, because otherwise we end up wasting a lot of energy arguing about what different people think is common sense, as opposed to taking an evidence- based approach to problems like homelessness.

And many studies have found that problems of homelessness are best addressed with upstream interventions by providing housing, rather than waiting for unhoused people to have the kinds of problems that run up costs, such as emergency medical services, interventions by police and, jailing and incarceration. 

They all tend to be much more expensive than actually giving people housing in the first place.

A lot of people feel, you know, out of frustration with the current economic situation, that we shouldn’t be rewarding people for not earning their way in the world. And yet we end up rewarding prison corporations for putting people in prison instead of providing the housing that would make a difference and better outcomes.

And it all makes sense to me to try to apply the solutions upstream at the root cause of the problems, because I was trained as a software engineer, as my background prior to having an interest in running for Congress, because we were trained that software bugs are most easily fixed upstream.

If the developer catches it and fixes it before it gets into the product, it’s at a cost of, say, $1. If the tester inside the organization finds the bug and refers it back to the developer to be fixed, it’s a cost of $10. Whereas if it actually ships as part of the product and isn’t noticed until a user runs across it, it costs $100. Right? So that’s just a principle. So it intuitively makes sense to me.

Different people have different careers where they have similar kinds of principles that they could probably describe that apply to the upstream-root-cause approach to problems.

I think because we have so much economic inequality within our society right now, many people are very frustrated and angry, and we have to be very careful about how those kinds of emotions can lead to a desire for punitive, punishing kinds of policies, to hurt people, and in some cases with ICE, for example, even kill people rather than take a reasoned, problem-solving approach to issues.

It’s a slippery slope. But if we can keep our heads about us and take a rational root cause approach to these problems, we can find solutions that are lower cost and apply those solutions and achieve better outcomes. 

Presenter: We contacted Marc to ask for his thoughts about a homelessness roundtable conducted by another candidate. 

Marc Baber (We The People Party): I did get a chance to talk briefly with Monique DeSpain after the Libertarian (Party) conference in Portland. I suggested that, you know, maybe a different follow-up meeting would be good with some of the behavioral science kinds of approaches, where there are some people at the University of Oregon who’ve done research on homelessness and the unhoused and costs and so on.

And these tend to confirm studies that have been done in other parts of the country, like Orange County, where the costs of housing people are much lower than the result of letting them remain unhoused, in terms of the cost of police interventions, prison time, emergency medical services. 

And so I offered to help set up another follow-up meeting with some of these university people and what may be some better solutions for that, not just with Monique, but with all politicians who are interested in applying this upstream solution approach.

But, across the board, a behavioral science approach just makes so much more sense than having people just arguing about what their gut tells them is the right solution, you know. And because different people’s guts say different things, right? 

A scientific approach to testing different solutions to see which works best is perhaps the most productive way to approach these problems on a large scale because when you make a decision in government, it affects thousands and millions of people sometimes.

So it makes sense to spend some money up front to get the right solution before you run off and start implementing something.

Presenter: He expressed concern about the turn towards punishment exhibited by the Trump-led Republican Party. Marc Baber:

Marc Baber (We The People Party): When I was growing up, we had LBJ’s Great Society and the question was: ‘What do we do to best guide people to get the education they need, get the medical care they need, so that everybody can achieve their full potential?’

And that seems to have been replaced with, ‘How can we beat people up to keep them from doing the things that we don’t want them to do?’ 

That’s sort of like the old office cubicle poster that said, ‘The beatings will continue until morale improves.’ Right? And that’s just not—it’s just not an approach that works in a human world, you know?

People respond more to compassion and encouragement rather than wrong making and punishment. 

The Trump-oriented or the Republican approach reflects the kind of punitive approach that’s become popularized in many different sectors, not just the unhoused, but also immigration and workers’ rights. 

All kinds of things seem to be getting the standard response that there should be more enforcement action and sometimes even violent enforcement action.

So it’s generating a lot of fear and uncertainty among the populations that are targeted. It’s part of a larger pattern of othering and dehumanization that precedes the withdrawal of support from certain sectors of society.

And I don’t want to dive too heavily into WWII analogies, but, you know, we’ve seen past history where societies have decided that, ‘Oh, we’d all be better off without this sector of our community,’ and what that leads to, and it’s tragic. And it’s something that hopefully we can avoid repeating again.

Presenter: Marc suspended his campaign this month, after the We The People Party lost ballot access. He said his campaign hoped to promote discussions of artificial intelligence and data centers. Marc Baber:

Marc Baber (We The People Party): These are big issues, the data centers, and where all of our water and electric resources are going. I would make sure that the costs would be such that the industry would be subsidizing people’s families’ electrical usage, rather than the other way around. 

We shouldn’t have the case where the average ratepayer is subsidizing the electrical usage of these data centers. 

Artificial intelligence is going to be impacting us in so many ways in terms of how wars are conducted, how the economy goes, with how many jobs are automated.

And I think a lot more needs to be said about that and very quickly, because that’s a problem that’s proceeding not at the pace of human beings, but at the pace of computer processors. 

We need to start doing some commonsense things, like making it illegal to put out videos that are fakes of people doing things that they never did, making it illegal to put out videos or audios.

One thing that really annoys me so much is to get through a YouTube of some famous person speaking and then realize, because of some anomalous mention of technology that didn’t exist during the lifetime of the person I believe is speaking—like Alan Watts referring to scrolling through social media, when he passed in 1973—I really wish this had been labeled as somebody else’s words before I started taking it in as if it had been Alan Watts, speaking at some talk.

So I think we need laws to label things that are produced by AI so that we can avoid them if we choose to. We require labeling on foods that go into our bodies and these are kind of foods that go into our minds. And we should at least be able to upfront make decisions about what we want to take into our minds.

But that’s just the first level. I think we need to be very clear about making sure that human beings are liable for the actions that their AI systems take.

For example, a top-level manager was doing a check-in with his HR staff to find out how they’d been applying AI, and they said, ‘We’ve reduced our time to hire by 60% using AI.’ 

And he said, ‘Well, let’s look at that.’ And he found that anybody that had a resume that listed a graduation date from college that was 2010 or before was 100% not hired by the AI system, that was instructed in rather vague terms to find candidates with ‘youthful energy.’

And he said, ‘You know, we could be liable for age discrimination with what this AI program is doing.’

And the HR manager said, ‘Well, I don’t see how that could be true because it’s the AI making the decision, not us.’

And he says, ‘The robot defense does not apply according to federal law.’

You can’t just say, ‘Oh, it’s not my fault, the robot did it.’ That is not a valid defense.

And we need to make sure that those kinds of rules where people are held responsible for what their AI systems do, and people and companies are held responsible, so that we don’t just get a bunch of robots violating laws and people getting off scot-free and people being harmed by not getting hired or not being accepted for programs because the AI makes a decision in some rather vague and nebulous way, but statistically is provable discrimination or other violations of the law.

And of course, we’re seeing AI used in the military for tracking and targeting of human beings, generating kill lists, controlling drones and so on, over in the Middle East and in Ukraine.

And these are other systems where we need to make sure that there are humans in the loop. We might be able to make certain exceptions for defensive things like if there’s an incoming hypersonic missile and there’s no time for a human to make a decision, we can automatically target that incoming missile and destroy it.

But maybe make rules about nothing traveling slower than 200 miles an hour or, nothing that appears to be a passenger plane and so on. So there could be rules established, under what conditions a human is required in the loop.

It’s far better to err on the side of having humans involved 99% of the time rather than, ‘Oh, well, you know, 70% of the time the AI seems to get it right.’  The margin for error as the AI systems have more and more impact on real world resources has to be brought closer to zero.

It’s like with self-driving cars. My first thought about self-driving cars was like, ‘Oh, no, we should never allow that,’ you know? And yet over time, we see that some self-driving cars, some software, some specific cars are able to have lower rates of automobile accidents than human drivers on average.

And that’s a scenario in which we might consider making the AI legal. Especially if, say, they prove to be 10 times safer than the average human driver, you could say, ‘Yeah, okay, that’s good enough. Let’s go with that.’

So there’s lots of use cases to be considered and lots of potential problems to be avoided. But also a lot of benefits that may accrue to society as a whole.

I would want to make policy so that as more and more work is automated, the benefits flow more to society as a whole, rather than just to the billionaire owners of the robot and AI companies.

And surprisingly, even J.D. Vance said something similar in which he recently argued that the AI companies should be 10% or 50% owned by the government, which would make us in control of what happens to their revenues and who guides the development of those products.

That’s really a huge kind of a shift that you would have a Republican proposing a policy that, on the face of it, is quite socialist. And yet this is what we’re going to have to balance as we go forward with this kind of technology is, they are the means of production, and we have to be careful who owns the means of production.

Because that ownership is essentially creating a monarchy for the billionaires who own the companies. And we want a democracy, so we need to choose mechanisms that share the power one way or another.

Presenter: That’s Marc Baber, of the We The People Party, former candidate for Oregon’s 4th Congressional District. You can sign up for his email list at MarcBaber.com.