157. Communicating the Future: How to Define Where We Want AI to Take Us

In communication, the ability to connect and to explain where you are and where you need to go is critical.

My name is Matt Abrahams, and I teach strategic communication at Stanford Graduate School of Business.

Welcome to Think Fast, Talk Smart, the podcast.

[MUSIC] Today I'm excited to speak with Russ Altman.

Russ is a professor of bioengineering at Stanford University.

His primary research interests are in the application of AI and data science to problems relevant to medicine.

Russ is also the host of Stanford Engineering's The Future of Everything podcast.

Welcome Russ.

I'm a big fan of your podcast and really enjoyed the opportunity we had earlier today where I was a guest on your show.

>> Thanks so much.

Thanks for having me.

I share enthusiasm for this podcast and it's kind of a thrill, I think it might be my first podcast as a guest and it couldn't be a better one.

>> Well, thank you and I'm excited for what we're about to do.

You ready to get started?

>> Yes.

>> Excellent, each of your podcast episode titles starts with the future of.

So let me start by asking you for your thoughts on the future of communication, especially how AI plays out in it.

>> So first of all, I'm glad you mentioned that I had you on my podcast this morning, because the answer is, you just told me the future of communication, but I actually thought about it before we chatted.

I'm an AI enthusiast.

I use AI in my research.

I'm part of the human centered AI Institute at Stanford.

AI in general has exposed lots of things that were hidden in our society.

So a lot of people hear about bias and unfairness and that is a problem with AI.

On the other hand, it was a problem before and in some ways, we need to give AI credit for exposing the unfairness and giving us something that we can work on to try to improve.

And the reason I mention this is I think for communication, in some ways, it's the same answer.

If you're worried that ChatGPT type tool can replace you, I think you need to think about why am I communicating?

What am I trying to say?

Am I being authentic?

Do I have a message?

Because really if those things are true, it shouldn't be a problem.

It should actually just help you amplify and improve your message.

So I think that we're at a time now where people have to ask hard questions about why they're communicating and how they communicate.

And I think that the AI is bringing this issues to the floor that were kind of insidiously in the background.

So you can see I'm an AI optimist and I think this will help.

>> I also think AI can bring a lot of value to communication.

And I really like your response that the thing to worry about is not AI taking over communications, the thing to worry about is, do I have a clear goal?

Do I know what is important to say and how can I craft it in a way to be helpful?

And I think AI can be an assistant in it.

For example, I have many non-native speaking students.

AI is a wonderful tool to help them learn some of the vernacular and learn other options for ways of saying things.

So I appreciate that perspective.

>> You know what?

It is interesting how good it is when you say, please explain this at the level of an eighth grader or at the level of a high school student.

And I think that's something we all can get better at, because listening, understanding who your audience is, and then making sure you're delivering at the right level is always a challenge.

And that seems to be one thing that it's kind of good at.

>> Absolutely and you can actually use it to flag jargon and terms that others might not understand.

You mentioned that you happen to use it in your communication.

Do you mind sharing some of the ways you use it?

>> My favorite story, if I may.

>> Sure. >> Because I had written a grant proposal, and I ran the one page summary of the grant through the large language model, and I said, please summarize the strengths and weaknesses of this summary of my research that I'm about to submit to the National Institutes of Health for research dollars.

And it came back and it had a few pretty good positives and negatives.

But my favorite negative is it said, you seem to be leaning on prior accomplishments too heavily.

And it was absolutely right.

So I've used it as a pre-human screen, before I go to my colleagues even, even my trusted colleagues, let's just see what the LLM has to say.

Maybe I'll fix it up and then give it to my buddy down the hall to take a look.

>> That's fantastic that it was helpful to you.

And I like this idea of it's actually a pre-editor.

>> Yes.

>> Did it recommend you need to see some therapy about some narcissism or no, no, no, it didn't.

>> That goes without statement.

But if I had asked, I'm sure it would have added to the chorus.

>> [LAUGH] I want to pull this thread a little farther about grants.

You have been incredibly successful at raising money for the projects that you do through the grants that you apply for.

Many of our listeners have to write proposals asking for funding from their bosses or from external organizations.

What guidance do you have when it comes to drafting proposals and grants?

>> I am passionate about this.

In general, the first thing you have to do is understand the mission of the person or institution to whom you are making a request.

Because if you don't hit that mission on directly, it's end of the game.

Given that, you need to have a big problem that you want to be part of the solution of and then a focused problem that you are actually going to provide the solution to.

You have to show that that subproblem is not solved.

Then you have to make the case that there's an opportunity that you and maybe others, but certainly you have.

And then you have to boil it down to three to five action items, followed by a summary of the state of the world when you're done.

And I believe that that formula can be used from everything from a grant to a federal funding agency to a marriage proposal.

And I've tried it in all different settings, and it really is the fundamental structure of any proposal.

>> When we turn off the microphones, I want to hear the future state that you used to propose to your wife, [LAUGH] but we'll- >> Happiness.

No, it's an easy one.

Honey, you will be happier then than you are now.

>> Excellent, so the key takeaway from what you just said, I think, is not only do you have to know your audience, what's important for them, and that's critical, regardless if it's a proposal or a pitch or a presentation.

But this notion of where we are to where we're going to from to to, is so important in pitches and presentations, and it sounds like in types of proposals.

And I like this notion of saying, hey there's this larger problem, and I'm going to focus on this subset.

And my unique value or my team's unique value is what's going to help us solve it.

Here's the specific things and then here's what the world's going to look like.

>> Thank you for listening Matt.

>> What you did just there is something I'd also like to ask you about.

Your work in the academic world is incredibly complex, is incredibly detailed, and yet you're so good at explaining and making that complex information accessible.

What are your thoughts on how to do that?

I know you do it strategically.

I know from the conversation you and I had earlier, you like using the whiteboard.

What are the things you think about when you take complex information to make it more accessible to people who just don't know as much without dumbing it down?

>> You know this very well that stories of what people remember, stories of what people like it's actually frustrating 'cause part of my work is statistics and statistics is really much more robust than a story.

We always say, you might have one good anecdote and that doesn't replace lots of actual data.

But I'm afraid that sometimes, a lot of times a good story does stand in for a lot of data because it's so compelling to the human mind.

So we're always looking for stories.

Analogy is a key part of both stopping the conversation and letting people catch up, but also saying, okay, wait a minute, let me rephrase what you just said, but let me not use the words you just said.

Let me map it onto an everyday situation like going shopping.

So you're telling me that the Internet and the way it works is like a very complicated mail system or whatever, and so can you listen to what the expert is saying and see an analogy to an everyday life thing?

And you're checking your understanding but on the, >> The chance that you're right, you're also helping the entire listenership or readership understand what was just said.

So that's important.

Related to that, and you already said it earlier, you're always have an antenna up for jargon and you, I'm sure you do this too.

You have to be willing to stop the conversation cold and say, you just said NIH, that's the National Institutes of Health.

They fund research in America.

And so part of what I'm doing always is a little antenna, listening for words that need to be defined or abbreviations that might not be commonly understood.

So I think analogy stories, and then adjusting, adjusting the levels.

So sometimes when you're talking with someone, you just have to send them signals in real time that we are talking at a level that requires a PhD and I'm going for high school college students here or whatever.

And then it's practice, and one of the great things I love both about this podcast and about your book is you don't say that you either got it or you don't got it.

All of this can be practiced and learned >> Absolutely that's a big point I like to try to make and encourage people and you are helping all of us learn this.

Something that I think can also be helpful, I'd love to get your input on this is there's work you can do in advance so you can actually stockpile some stories.

Or think about some potential analogies that might work so it doesn't all have to happen in the moment.

You can pull them in and at the same time, some pre-work you can do is you can prepare the audience.

So in a meeting invite, for example, you could say, and we're going to be talking about concepts like this.

So people come in primed which I think makes your job easier.

What are your thoughts on that kind of pre-work.

>> I totally agree I had one more point written down, and I just had decided that I had talked too long a second ago.

But I said, anticipate the tough spots and validate that they're tough.

So say to the audience, this guy's talking about quantum physics, this is hard stuff.

Here's how we're going to try to manage through it together.

So I just want to say that you're absolutely right.

I think the key idea of prep for an interview or for creating a document or any kind of communication is anticipating the tough moments, the tough concepts, and exactly as you say saying, okay, I'm going to try to fly with this analogy.

I think this'll work, I think it'll make it more accessible, more relatable.

And that's part of doing your homework and it's not that hard to just think.

I spend before, I talk to you before I talk to anybody, I spend the night before basically Google stalking you, looking at what have you said, what have you done.

Part of it is so I can have emergency topics to talk about if it turns out that you're reticent to speak.

That wasn't a problem.

>> [LAUGH] >> But the other, I've never been accused of that.

>> Right, exactly, nor have I?

However, the other thing was what might he say that might be surprising or troubling, and how am I going to manage that?

And then you go in with this confidence that kind of whatever comes down the pike, we're ready for it.

And you may use them or you may not use them, but having them just has you, this sense of preparedness.

>> Right, and that helps you be confident and more present because you're not reaching ahead thinking what's coming.

You do a good job of preparing to ask the questions you ask on your podcast.

I saw that firsthand, I've listened to the show and see how you do that.

That preparation in advance, although I know that I'm recommending everybody cyber stalk people, but I do think it's important to do your homework.

What advice do you have for asking questions that can really help focus and guide communication and conversations?

>> Specificity in the question is key.

Specificity rules in the world, in my opinion.

People want to hear stories that are specific, not general, and they want to hear questions that are specific.

But then my second tip helps take a little bit of pressure off, because I think the follow-up question is arguably more important than the initial question.

because the initial question gets the listeners and the guests, if we're taking about a podcast.

>> Sure. >> Into a certain space.

And there might be some hemming and hawing as they gather their thoughts, but then they say something, and now they're warmed up.

And the audience is warmed up.

And now you can come up with a follow up question that might be the meaty question.

And I've heard people who are very good who do interviews, David Letterman, Johnny Carson, if you watch their follow ups, were always very good.

>> In addition to being specific and being open to asking follow up questions.

The use of questions isn't just for people who do interviewing.

You can be very effective in a meeting and giving feedback through asking focus specific questions and in some ways, I think people receive that in a more open way than if I give you some declarative statement.

>> Yes, I think you're absolutely right.

I notice that students, like my PhD students, the last thing that they get good at before they're ready to launch and graduate is answering questions after a seminar.

Because they come to it thinking that questions are an attack.

And so they get defensive, their answers are almost off-putting.

And so there's a mentorship that you do to say, you need to consider every question as a little gift.

>> I love it. >> Yeah,

it's a gift it shows I listened to what you just said.

I've processed it, and now I actually want to have a conversation with you that like, it doesn't get any better than that.

So yes, I'm a big fan that questions can be powerful.

They can say thank you.

They can say, I heard you, and they can say, and I have a little bit of an opinion here without even saying it as I have a little bit of an opinion here.

>> Right, seeing questions and actually lots of spontaneous situations like giving feedback and even introducing somebody as opportunities rather than threats and challenges is a fundamental mindset shift.

And I love that you spend time with your students, helping them get there because many of it's instinctual when somebody comes up with a question we get defensive.

>> Yeah. >> But

if we can see it as an opportunity to connect to learn to collaborate, I think is wonderful.

>> My best friend is a sports writer and he's and but really more importantly, he's a journalist.

He's famous for making his points with these very understated beautiful questions where he doesn't even have to follow up and tell you what he really thinks because it's obvious from his question.

And I think to myself, what a great journalist he is and was because of this skill.

>> We had a whole episode with Deborah Schifrin on how to ask good questions.

She said to pause, so after you ask the question pause and after the person answers, pause even longer because more comes out.

>> I'd love to move to another area where you are an expert.

You work on highly complex topics, but you also work in highly complex teams with people spread out all over the place with different areas of expertise in different disciplines.

What best practices do you use to foster creativity and collaboration while trying to avoid destructive conflict and repetitive actions and wastes of time?

>> It has become clear that team science is the most effective and is the most exciting in many ways.

You must understand and acknowledge the rewards and motivations of your collaborators.

Because they're unlikely to be identical to your own.

And if you leave that unstated or worse, not understood, it's just a recipe for a lot of wasted energy and wasted words.

And that leads to number two, which is you have to build the relationship, the human relationship separate from the scientific or business collaboration.

And those two go together, understanding and building a relationship.

But that time, which is kind of a pain in the neck because it means you have to have lunch with them and maybe you don't want to have lunch.

Maybe you want to just take five minutes to shove food down your throat because you're busy.

But the time to build that relationship pays off in spades.

So those two things come first.

Then you have to, I think diversity and teamwork has become, there's now literal evidence.

That if everybody looks and sounds like you and has the same training and the same perspective and the same goals and motivations, as I've talked about a moment ago, that is not the recipe for the best team.

So you have to take a hard look.

At who've you have invited to the room.

And if you're really brave and courageous, you'll say this is a good group but this is not the right group for this project.

And we either need to add to it, or even more difficult, I need to make some substitution, which can be a whole set of issues.

And then it's about roles and responsibilities.

And with the relationship in place and with an understanding of the reward systems and with a diversity, now you can say, finally okay here's what we're trying to do.

How are we going to break this down and how are we going to do roles and responsibilities so that people feel like they're part of a team that is functioning?

And then obviously, and I could not be on this podcast without saying, having mechanisms for both formal and informal communication in place, and understood.

>> Alot of richness in that answer.

>> Sorry.

>> No no, no, it's all good.

The fact that you first have to get alignment on goals and expectations is critical, forming those relationships which then allow of what you've said and I will just put an exclamation point on this notion of diversity.

Diversity of opinions and experience and goals is what brings around the richness but also requires that you have to define the roles, responsibilities and ways of communicating because if people have those differences, they bring that different way of communicating too.

I want to drill down on one aspect that I was very specific in the question I asked.

I said avoiding destructive conflict.

I know in the work you do that encouraging some disagreement and conflict is actually helpful to achieving the goals.

How do you thread that needle?

>> Where it comes up in my life literally every week is that all the students and postdocs in the lab, take turns every week giving what we call a group meeting.

It's a it's a research and progress talk about how they're going, so and this is what their colleagues, who they sit next to all week.

So what's supposed to happen?

What's supposed to happen is we're supposed to be supportive.

We wanted them to feel comfortable, but they're supposed to be practicing for an audience that may be hostile, like scientific colleagues, can be hostile.

And so I, as the leader of the group, have to figure out, how do I set this up so that they are getting the hardest questions, the hardest, most uncomfortable, scientific, technical questions from these people that they sit with every day, but it doesn't lead to a toxic environment.

Lack of trust, and people saying, I don't want to give my group meeting, I don't trust this group.

I have to set it up so that they can present, I can ask hard questions, but then we have a culture of thanking them for the talk, acknowledging the hard things that they've uncovered as hard, thanking them for the solutions that they've produced.

And if I see problems, I have other communication methods in the lab to try to identify those and deal with them outside of that forum.

But you're absolutely right.

This is like the number one thing for a team leader to be aware of because the rigor slash support tension is there all the time.

>> Russ, one of the things I love about hosting this podcast is the ability to just talk with amazing people, yourself included.

I always learn something.

What is something that you've learned from your many years of hosting The Future of Everything that you think might help the listeners of Think Fast Talk Smart?

>> The people who I interview, what they do is the thing that they think is the most important thing they could do.

So you talk to a civil engineer, it's not that this is one little interesting problem in civil engineering.

This is the most interesting problem in civil engineering to this person.

And indeed it's the most important thing they can think of doing.

And so the passion that's scientists and engineers, because that's what we focus on, although we have other guests as well.

But the passion of scientists and engineers in approaching their work and the magnitude of that passion probably shouldn't have surprised me, but it did.

And I learned that you need to get to people's passion to really understand what makes them tick.

>> And inviting that passion out through questions and through giving space for them to do that can really help you connect and build a stronger relationship.

So you don't have to be a podcast host to take advantage of that and that's one of the things I love too is just finding people's passions.

Sometimes people are so passionate about something I never thought of, and just their passion makes me interested.

>> And some people don't wear it on their sleeve and you have to work a little bit to get it out, but it's always there.

>> Yes.

Well, I know you listen to the podcast, so you know that before we end, I like to ask three questions.

One I create just for you.

>> My goodness.

This is so exciting.

[LAUGH] >> And two that are similar for everybody else I interview.

You up for yours?

>> Yeah, absolutely.

>> All right.

You are very future-focused, so what is one thing about the future that you're particularly excited about?

>> I am excited about young people.

Young people starting to form their lives and their direction without fail every year gives me hope for the future.

As you may have come to your attention, the world is currently facing a lot of problems, and it's very easy to get very negative.

And if you want to combat that negativity, be involved in an admissions process or a hiring process of young people like for their first job or for their next job.

And if you're not inspired, then I think you need to do some deep introspection at.

>> I love that answer simply because I have lots of friends who are getting curmudgeonly and they will say, these young people, what's this world going to amount to?

And I say, hey, it's us that's causing the problems because if you come in and see my students or the people I coach, there's a lot of hope.

And there's a lot of excitement.

And I'm glad you echoed that because I say that all the time.

Question number two, who is a communicator that you admire and why?

>> My dad.

>> Tell me why.

>> He had a great voice, much better than mine.

He had an FM voice, and he had a great way with words, and he did lots of the things that you discuss all the time.

When I'm trying to do my best at communicating, I'm channeling dad.

>> So I actually am getting a little choked up because I feel exactly the same way about my father.

My father had an amazing resonant voice, was a a great communicator.

Get to the point.

What's the point?

And and it really helped, me try to be more focused and concrete.

>> So be if I may.

>> Yeah. >> Between the podcast we did this morning and then we're doing now, I gave a talk and somebody said, what's the best piece of advice you ever got?

And I said it was from my dad and it was get to the point.

>> How funny.

Question number three, what are the first three ingredients that go into a successful communication recipe.

>> You need to know who you're talking to.

You need to know the audience and a little bit about their goals, hopes, and dreams. Second, you need to be specific in your message.

And then the third one I'm going to go with, make sure you try to keep it fun, light, and enjoyable.

Life is short, and if you can do something with a sense of joy, why not do it with a sense of joy?

So, even if it's at the most mundane message, think about how can I make this a positive experience for me and for the person who's about to listen to me and have a conversation with me.

So those are my three.

>> This notion of connection and engagement through the energy you bring to your topic is really important.

There are people who study things that I don't understand and prior to hearing them talk about it or write about it, I didn't care about.

But their passion and their ability to connect it to me and make it engaging makes a big difference.

>> I agree.

>> Russ, thank you so much.

You did an amazing job of educating us and keeping it engaging and interesting.

I appreciate it.

All of us can learn from you in terms of how to connect with our audiences, manage some of the tricky situations through questions and in structuring our environment so that we can take the advantage of conflict.

I appreciate your time.

Best of luck on the podcast.

Thank you.

>> Thank you very much.

I really enjoyed this and congratulations on your successes.

>> Thank you.

You've been listening to Think Fast Talk Smart, the podcast.

If you've enjoyed this episode, you might want to check out similar topics discussed in Episode 109 with Francis Frey, how to communicate compolex ideas simply and effectively.

And Episode three with Lauren Weinstein, whe knowing too much can hurt your communication, how to make complex ideas accessible.

Our show is produced by Michael Reilly, Jenny Luna, H.,

Ash, and me, Matt Abrahams, with help from Podium Podcast Company.

Our music is by Floyd Wonder.

Please find us on YouTube and wherever you get your podcasts.

Be sure to subscribe and rate us.

Also follow us on LinkedIn and Instagram and check out fastersmarter.io, for deep dive videos, English language learning content and our newsletter.

[MUSIC]