Machine values with Joi Ito and Barack Obama

I read a fantastic conversation on Wired magazine between Joi Ito (Director of the MIT Media Lab) and Barack Obama yesterday. The conversation was about artificial intelligence, self driving cars and the future of technology. I would strongly recommend reading all of it. However, if you are out of time, here is an important excerpt from the discussion.

JOI ITO: This may upset some of my students at MIT, but one of my concerns is that it’s been a predominately male gang of kids, mostly white, who are building the core computer science around AI, and they’re more comfortable talking to computers than to human beings. A lot of them feel that if they could just make that science-fiction, generalized AI, we wouldn’t have to worry about all the messy stuff like politics and society. They think machines will just figure it all out for us.

OBAMA: Right.

ITO: But they underestimate the difficulties, and I feel like this is the year that artificial intelligence becomes more than just a computer science problem. Everybody needs to understand that how AI behaves is important. In the Media Lab we use the term extended intelligence (Extended intelligence is using machine learning to extend the abilities of human intelligence). Because the question is, how do we build societal values into AI?

OBAMA: When we had lunch a while back, Joi used the example of self-driving cars. The technology is essentially here. We have machines that can make a bunch of quick decisions that could drastically reduce traffic fatalities, drastically improve the efficiency of our transpor­tation grid, and help solve things like carbon emissions that are causing the warming of the planet. But Joi made a very elegant point, which is, what are the values that we’re going to embed in the cars? There are gonna be a bunch of choices that you have to make, the classic problem being: If the car is driving, you can swerve to avoid hitting a pedestrian, but then you might hit a wall and kill yourself. It’s a moral decision, and who’s setting up those rules?

ITO: When we did the car trolley problem (The car trolley problem is a 2016 MIT Media Lab study in which respondents weighed certain lose-lose situations facing a driverless car. E.g., is it better for five passengers to die so that five pedestrians can live, or is it better for the passengers to live while the pedestrians die?), we found that most people liked the idea that the driver and the passengers could be sacrificed to save many people. They also said they would never buy a self-driving car. [Laughs.]

DADICH: As we start to get into these ethical questions, what is the role of government?

OBAMA: The way I’ve been thinking about the regulatory structure as AI emerges is that, early in a technology, a thousand flowers should bloom. And the government should add a relatively light touch, investing heavily in research and making sure there’s a conversation between basic research and applied research. As technologies emerge and mature, then figuring out how they get incorporated into existing regulatory structures becomes a tougher problem, and the govern­ment needs to be involved a little bit more. Not always to force the new technology into the square peg that exists but to make sure the regulations reflect a broad base set of values. Otherwise, we may find that it’s disadvantaging certain people or certain groups.

I think this is a critical conversation – one that we must all have as we build toward the future.

Machines are not sources of disembodied truth. Anyone who has conducted any kind of analysis with huge data sets will tell you that. Machines take with them our assumptions and judgment. Similarly, artificial intelligence isn’t going to conjure up values. We will teach AI to make these decisions. The self driving car decision is just one such example. President Obama remarks later that there aren’t enough people thinking about “the singularity.”

That is true. Most of us are wrapped in day to day nonsense that isn’t really going to matter in the big scheme of things. As technology becomes a bigger of our lives, the onus is on us to make sure we have discussions on how we build this technology. Machine values are not going to save us.

Human values are.

PS: How many heads of state can you imagine having such a thoughtful conversation about the future?

Deadliest animals and availability bias

Availability bias is a mental shortcut we use when evaluating a topic or decision by relying on immediate examples that come to mind. For example, below is a chart with the planet’s deadliest animals in 2015.

Before you look at it, what animals come to mind when you think of deadly animals?

I think I might have had lions/tigers, snakes and sharks in some order.

The deadliest animals were mosquitoes and humans by a distance. The shark attack number is the lowest of the lot. But, I was drawing from the fact that I have seen more pieces of news about Shark attacks than Jellyfish killings. And, that is a classic example of Availability bias.

Similarly, the world is more peaceful than it has ever been. And, yet, politicians and the media would love to have us believe that we are in constant danger. Hearing something regularly doesn’t make it true.

Thanks to my good friend Bill Gates for sharing this graphic (just kidding of course). And, congratulations to the Gates foundation for leading the charge in the fight against malaria and contributing to the decline of Malaria by 57% in sub Saharan Africa.

The hard thing about self worth

The hard thing about self worth is the presence of the word “worth.” Its presence primes us to look for numbers and comparisons around us. It feels natural to look at the races we’ve run and won, our place in the pecking order in our careers, the size of our car or house, and so on.

But, spending our time on numbers and comparisons is just a race to the bottom. There is just more unhappiness and insecurity to be found there. Everybody loses.

The challenge with self worth is measuring ourselves by who we are being today versus who we were yesterday. Are we living better? Have we been behaving in a manner consistent with our values? Are we taking good care of ourselves?

Again, that’s hard to do. After all, any kind of measurement prompts us to look outward.

Perhaps the biggest clue is in the word that comes first. Self worth is more about the self than the worth.

People leverage and System leverage

Leverage means using something to maximum advantage outside of the financial world. It is often used to describe human capital. For example, new hires in a company ideally provide leverage to their managers. And, supporting functions provide leverage to their sales teams.

I find two kinds of leverage in organizations – people leverage and system leverage. The underlying concept is similar. Leverage provides for a stronger support system for execution. However, while people leverage focuses on people to provide the support system, system leverage relies on processes and systems.

Imagine you are the one woman customer service center specialist. Your company is growing quickly and you decide to hire someone. The guy you just hired provides you immediate leverage. He takes all the basic stuff off your plate and allows you to focus on more strategic stuff. Soon, you could expand this to a team of three. This is a classic example of people leverage.

However, let’s assume your first hire does a little more than you asked him to do and creates a really good FAQ resource for your customers. All of a sudden, you may not need to hire three people. That resource has helped provide system leverage. It allows you to operate at a higher level without adding people to the organization to solve the problem.

There are a couple of important takeaways once we understand this difference. First, most organizations intuitively understand people leverage. However, there aren’t enough that get system leverage. The best organizations and teams have fantastic processes and systems that enable their people to perform at a high level. This is often what makes large corporations tick. There are many large corporations whose human capital potential are definitely not being utilized. However, thanks to the strength of their systems, they still deliver impressive results. Of course, the truly great corporations have both.

Second, when you and I are hired to a new job, we provide automatic human leverage. We might even provide our manager the leverage created by two hires if we were very good. However, there is no better multiplier than when we build systems. Looking for inefficiencies in how we operate and solving them by putting systems, tools and processes in place is among the highest impact things we will do.

Not okay

In case you missed it, Donald Trump brushed his lewd 2005 tape as “locker room talk.” The Trump campaign seemed to imply that it is normal for men to discuss assaulting women in locker rooms. No, that’s not true and saying so is not okay. The campaign does live in its own alternative reality, however. So, perhaps that shouldn’t surprise us. Here are six lessons I took away from the weekend –

1. Twitter was on fire this weekend. There’s a dark side to Twitter and that’s a worthy subject of a different blog post. However, during a weekend when there’s so much news and commentary, Twitter is by far the most interesting place to be. I rarely use Twitter. But, this weekend, I was checking my Twitter feed every few hours.

2. There were a couple of very powerful threads on Twitter. One of them was when author Kelly Oxford asked women to share their sexual assault stories. A million women shared their stories. It was heartrending.

3. While the tape caused great damage to the Trump campaign, FiveThirtyEight shared that 10% of Trump supporters actually had a more positive view of the candidate after the tape went out. As Nate Silver put it, this is something else.

4. Muslims lit up Twitter with #MuslimsReportStuff last night. Human creativity and humor are beautiful things.

5. I could sense the false equivalence as I watched the debate yesterday. I could feel myself hoping Hillary came out as warm, inspiring and coherent. As long as Trump didn’t commit murder on screen, he’d probably do okay. Not okay.

6. Back to the women issue, it was nice to see Republican leaders denounce it. However, they all spoke about the fact that they had wives, daughters and mothers to whom this was insulting. Don’t they know any Mexican immigrants? Muslims? What about an example for their sons?
It frustrates me when people keep associating sexual assaults with their daughters. Sure, their daughters are the ones who are affected. But, it is their sons who could either be indulging such talk in said locker rooms or denouncing it.

I decided to cap myself at six. The list is long.

A month or so ago, I had a related post up here. Someone in the comments asked about the idea of voting in protest. I mentioned then that I didn’t have a point of view. I do now. If you share similar values and world view, then Trump is scary. Then again, if you don’t, Hillary is a liar. I still find it hard to believe that you could think she’s a worse human being than Trump. That, I think is a result of unconscious gender bias. Either way, will all this change anyone’s mind or cause them to support Hillary while they were supporting Trump? Probably not. Most people have made up their mind. Their friends who share “news” on Facebook keeps reinforcing that view.

But, I think Clay Shirky’s assertion that “There’s no such thing as a protest vote” is spot on. There isn’t. That’s just the way the system is set up.

Not okay. But, as history shows, realizing that is hopefully a step toward making it better.

The London taxi driver study – The 200 words project

In a lengthy study, Eleanor Maguire and Katherine Woollett from the neuroimaging center at University College London followed a group of 79 trainee taxi drivers and 31 controls (people who weren’t in training). Over time, they took snapshots of their brain structure using MRI and studied their performance on memory tasks.

The trainee taxi drivers had to memorize a map of London with all 25,000 streets and thousands of landmarks to pass; one of the toughest qualification tests in the world. As a result, only 39 of the trainee taxi drivers passed the test.

The researchers famously saw a greater volume of cells in the successful drivers’ hippocampus. This is the area of the brain associated with spatial memory. Over time, it also showed that the longer the driver’s experience, the larger the hippocampus. And, on the flip side, as time passed after a driver retired, the hippocampus shrunk to normal size.

This hippocampus study famously pushed us to consider the hypothesis that our brains develop with exercise and are not “fixed” as was previously assumed.

So, all this leads us to a big question we’ll tackle next week. Did the taxi drivers who passed have some innate talent or genetic predisposition that enabled them to pass the test?

The human brain remains ‘plastic‘ even in adult life, allowing it to adapt when we learn new tasks. – Eleanor Maguire

—
Source and thanks to: Peak by Anders Ericsson, The Hippocampus study, Wired’s article on the study

(The 200 words project involves sharing a story from a book/blog/article I’ve read within 200 words)

Create dots to connect

The famous Steve Jobs commencement speech at Stanford made the line “The dots always connect backward” famous. I love the line and have found it to be true. There’s just one thing I’d add – we have a responsibility to create dots today that we will be able to connect tomorrow.

I have thought about this line many a time over the past few years. There are so many instances when I’ve looked back at a piece of work I did ages ago and have seen it pay forward in ways I would never have imagined. This has especially been the case with side projects. I learnt video editing for a talk show side project in university. Naturally, I became the de-facto creative video editor in the first couple of years at the consulting firm I worked with. I tried reaching out to authors and venture capitalists I admired and have continued to learn tremendously from those relationships. There are many examples at work – the skills I’ve picked up in the past have nearly always come of use.

So, yes, the dots always connect.

But, phrasing it that way seems to let us off the hook just a bit. It seems to absolve us of the responsibility to do something about it now. And, I think that’s because of the implicit assumption – that we have taken the responsibility to create the dots today that we will connect tomorrow.

So, enjoy the feeling of dots from the past connecting. But, every time you experience that wonderful feeling, create a new dot today that you’ll enjoy connecting tomorrow. Take initiative. Build something. Or invest in your learning and share that learning.

Create dots to connect.

One conversation decision to rule them all

There are so many decisions we need to make for conversations to go well. These decisions could range from listening to being constructive to being thoughtful about what we say. However, I’ve come to realize there is one conversation decision to rule them all.

And, that decision is whether or not to take it personally.

If we don’t explicitly make that decision, we might react defensively to everything that is being said. And, once we do that, it is very hard to recover. But, decide not to take it personally and we give ourselves the opportunity to listen, to understand, to be thoughtful and to respond constructively.

The key to making this decision is understanding that the issue is rarely about us. We might have contributed to it. But, it is unlikely it is about us unless we decide to make it so. The quicker we get out of the way, the sooner we can move to prompt resolution.

We have conversations very often. We might as well learn to do them well.

Back the car in at night

Take an extra minute as you park to back the car in at night. This way, you won’t have to spend any energy attempting to back it out in the morning. And, more importantly, you’ll reduce the chances of starting your day with a scratch on your car.

Similarly, decide on your priorities for tomorrow at the end of the day today. If possible, keep the list of things to do and the appointments ready to go.

Finally, if you are planning on exercising the morning, keep your exercise clothes ready next to your bag.

A few small things at the end of the day can make a big difference tomorrow. Done consistently, they can remove a ton of early morning friction.

A good start does not a successful day make. But, it does makes sure you’re giving yourself a good shot at it.

Bright eyed

An executive and former entrepreneur shared a trait he looked for when hiring people – bright eyed. To him, bright eyed stood for the desire to seek and welcome new experiences. It pointed to people who were eager to put learning above all else.

Benjamin Zander calls this shining eyes. He describes the goal of leadership to make sure you surround yourself with people who have shining eyes. He asks – how can we as leaders improve ourselves to inspire those around us?

Similarly, a friend once described himself as someone who enjoyed working with obsessive people who loved learning. He said he always learnt a ton when he worked with obsessive people because they cared so much. It didn’t really matter what they worked on.

The most beautiful part of these descriptions is that they don’t refer to some innate talent. After all, this just means caring tremendously about learning because you realize that it’s a very uplifting way to live this life. Learning enables us all to rise above the small things, focus on our growth and, in the process, help others grow to become the best version of themselves.

Bright eyed is a choice. And, with it, we don’t just choose a better path in our careers (we do that as well), we choose a better life.