WEBVTT - Greylock Partners Partner Reid Hoffman Talks AI Agents

0:00:02.520 --> 0:00:08.160
<v Speaker 1>Bloomberg Audio Studios, podcasts, radio news Now.

0:00:08.200 --> 0:00:11.280
<v Speaker 2>Earlier this month, an AI model called Manus went viral

0:00:11.360 --> 0:00:15.040
<v Speaker 2>for its apparent ability to act more independently than AI chatbox.

0:00:15.080 --> 0:00:15.240
<v Speaker 1>Now.

0:00:15.280 --> 0:00:19.040
<v Speaker 2>The development of the so called artificial intelligence agents have

0:00:19.200 --> 0:00:22.160
<v Speaker 2>raised concerns that they will erode human ability to think.

0:00:22.160 --> 0:00:25.240
<v Speaker 2>But Read Hoffman, the Lincoln co founder, Microsoft board member

0:00:25.480 --> 0:00:28.840
<v Speaker 2>and Greylock partner, thinks the opposite. Now He's just published

0:00:28.840 --> 0:00:31.800
<v Speaker 2>a book in which he basically argues that aisystems gain

0:00:31.920 --> 0:00:35.880
<v Speaker 2>greater abilities, they will enhance human agency, hence the book's

0:00:35.920 --> 0:00:38.479
<v Speaker 2>title Souperagency? What could possibly go.

0:00:38.560 --> 0:00:39.840
<v Speaker 1>Right with our AI?

0:00:39.920 --> 0:00:41.880
<v Speaker 2>Feature? And read is here with me? Thank you so much,

0:00:42.120 --> 0:00:43.920
<v Speaker 2>Read Hoffen for joining us. I mean, this is the

0:00:43.960 --> 0:00:45.680
<v Speaker 2>fresh of breath air because there is a lot of

0:00:45.720 --> 0:00:47.680
<v Speaker 2>concern and a lot of worry that AI takes over

0:00:48.320 --> 0:00:52.040
<v Speaker 2>the computers will be in charge and will basically stop

0:00:52.159 --> 0:00:53.280
<v Speaker 2>using our critical thinking.

0:00:53.479 --> 0:00:57.120
<v Speaker 1>Yes, but actually, in fact, if anyone plays with AI today,

0:00:57.320 --> 0:01:01.360
<v Speaker 1>it is the most amazing education technology.

0:01:01.040 --> 0:01:02.800
<v Speaker 3>We have created in human history.

0:01:03.040 --> 0:01:05.399
<v Speaker 1>If you want to learn anything, I use it to

0:01:05.480 --> 0:01:09.080
<v Speaker 1>learn everything from quantum mechanics to like, huh, I wonder

0:01:09.120 --> 0:01:12.640
<v Speaker 1>what cooking souved in this way looks like it's everything.

0:01:12.920 --> 0:01:15.520
<v Speaker 2>But I guess the concern is that, you know, especially

0:01:15.720 --> 0:01:18.399
<v Speaker 2>people that go into a first time job, or students

0:01:18.600 --> 0:01:21.640
<v Speaker 2>or like you know, college kids don't use their critical

0:01:21.640 --> 0:01:23.880
<v Speaker 2>thinking anymore. Because if you just go into chat box

0:01:24.200 --> 0:01:25.840
<v Speaker 2>and say write me a song, it with this, this,

0:01:25.880 --> 0:01:27.200
<v Speaker 2>and this, it does it for you.

0:01:27.440 --> 0:01:29.880
<v Speaker 1>Well, it definitely can do a bunch of things for you,

0:01:30.000 --> 0:01:32.360
<v Speaker 1>but that can help you elevate your game.

0:01:32.800 --> 0:01:33.000
<v Speaker 3>Right.

0:01:33.080 --> 0:01:35.160
<v Speaker 1>So it's a little bit like if you were just

0:01:35.240 --> 0:01:38.520
<v Speaker 1>copying Wikipedia as handing in your essay, sure you could

0:01:38.520 --> 0:01:41.600
<v Speaker 1>do that, but actually, in fact you should use it

0:01:41.640 --> 0:01:44.680
<v Speaker 1>to inspire you to make you think better, to say, hey,

0:01:44.880 --> 0:01:47.400
<v Speaker 1>like for example, when I was writing Superagency, I would

0:01:47.920 --> 0:01:50.600
<v Speaker 1>put in sections and say, how at a history of

0:01:50.680 --> 0:01:54.400
<v Speaker 1>technology specialists critique what I've said, and then I understand

0:01:54.400 --> 0:01:56.400
<v Speaker 1>it and I can decide whether or not I change

0:01:56.440 --> 0:01:59.080
<v Speaker 1>your editings or and therefore the book gets better.

0:01:59.200 --> 0:02:02.200
<v Speaker 2>So I read and reduce cognitive capabilities. Right, this is

0:02:02.240 --> 0:02:05.480
<v Speaker 2>a big concern that we stop thinking, that we think less,

0:02:05.480 --> 0:02:06.520
<v Speaker 2>that we think differently.

0:02:06.800 --> 0:02:09.639
<v Speaker 1>Well, I think it makes it just like all technology,

0:02:09.880 --> 0:02:12.800
<v Speaker 1>you can approach it being lazy, and so if you

0:02:12.880 --> 0:02:15.320
<v Speaker 1>just say, okay, I'm going to outsource it, just like

0:02:15.360 --> 0:02:17.800
<v Speaker 1>for example, I'm going to say, whatever the first search

0:02:17.800 --> 0:02:20.880
<v Speaker 1>result on Google is, that's the answer. And if you

0:02:21.000 --> 0:02:24.400
<v Speaker 1>do it that way, then of course that doesn't help

0:02:24.440 --> 0:02:27.320
<v Speaker 1>you extend. But if you do anything in terms of

0:02:28.280 --> 0:02:31.000
<v Speaker 1>having it be a dialogue with you, having it extend

0:02:31.040 --> 0:02:34.680
<v Speaker 1>your capabilities asking a question, getting an answer, asking another question,

0:02:35.280 --> 0:02:38.560
<v Speaker 1>then it it greatly amplifies your capabilities.

0:02:38.800 --> 0:02:41.000
<v Speaker 2>How do you get rid of the biases, because if

0:02:41.040 --> 0:02:45.399
<v Speaker 2>you have too many biases then it excus of course democracy.

0:02:46.000 --> 0:02:50.200
<v Speaker 1>Well, so all of the major AI labs are trying

0:02:50.240 --> 0:02:54.560
<v Speaker 1>to get it as kind of call it unbiased as possible. Now,

0:02:55.080 --> 0:02:57.919
<v Speaker 1>within human perspective and human knowledge, there's always some things

0:02:57.919 --> 0:02:58.280
<v Speaker 1>a bias.

0:02:58.280 --> 0:02:59.120
<v Speaker 3>We're always learning.

0:02:59.480 --> 0:03:02.280
<v Speaker 1>Like if we kind of look at human beings fifty

0:03:02.360 --> 0:03:04.280
<v Speaker 1>years ago and we are fifty years past and we

0:03:04.320 --> 0:03:06.280
<v Speaker 1>look at them and say, oh, they were biased about this,

0:03:06.680 --> 0:03:09.280
<v Speaker 1>I'm certain humans fifty years from now will be looking

0:03:09.320 --> 0:03:11.880
<v Speaker 1>at us the same way. So it's an ongoing process

0:03:12.240 --> 0:03:14.040
<v Speaker 1>with us as well as the technology.

0:03:14.480 --> 0:03:16.160
<v Speaker 2>Is there anything that worries you about AI?

0:03:16.600 --> 0:03:20.040
<v Speaker 1>The primary thing that worries me is I call AI

0:03:20.120 --> 0:03:23.800
<v Speaker 1>the cognitive industrial Revolution. It's both for the upside, which

0:03:23.840 --> 0:03:27.639
<v Speaker 1>is this whole society we live in, middle class education,

0:03:28.200 --> 0:03:32.639
<v Speaker 1>medicine all comes from the Industrial Revolution. That same amplification

0:03:32.800 --> 0:03:35.960
<v Speaker 1>is coming, but the transitions are difficult. So the thing

0:03:36.000 --> 0:03:38.840
<v Speaker 1>that primarily worries me is to say, look, we're going

0:03:38.880 --> 0:03:42.200
<v Speaker 1>to have to navigate this challenging transition, just like the

0:03:42.240 --> 0:03:44.360
<v Speaker 1>Industrial Revolution was a challenging transition.

0:03:44.880 --> 0:03:46.640
<v Speaker 3>But that's how we have.

0:03:46.760 --> 0:03:50.200
<v Speaker 1>Our children on our future generations be prosperous and have

0:03:50.480 --> 0:03:52.720
<v Speaker 1>amazing societies.

0:03:52.760 --> 0:03:54.560
<v Speaker 3>And so that's the challenge we need to rise to.

0:03:54.880 --> 0:03:58.520
<v Speaker 2>So what's the right way of either designing AI or

0:03:58.560 --> 0:04:00.920
<v Speaker 2>designing safeguards for AI?

0:04:01.120 --> 0:04:05.320
<v Speaker 1>Yeah, well, part of it, there's kind of a two

0:04:05.400 --> 0:04:08.280
<v Speaker 1>part audience for super agency. One is the people who

0:04:08.320 --> 0:04:12.680
<v Speaker 1>are AI fearful or concerned to help them become AI curious.

0:04:12.920 --> 0:04:17.320
<v Speaker 1>But it's also for technologists, which is design for human agency,

0:04:17.400 --> 0:04:19.280
<v Speaker 1>designed for increasing human agency.

0:04:19.600 --> 0:04:20.480
<v Speaker 3>That should be your.

0:04:20.360 --> 0:04:24.520
<v Speaker 1>Design principle and fundamental And the book is also I

0:04:24.520 --> 0:04:25.520
<v Speaker 1>hope helpful for them.

0:04:25.760 --> 0:04:29.240
<v Speaker 2>So read this is basically putting the human you know,

0:04:29.560 --> 0:04:31.800
<v Speaker 2>at the center still of everything. So how they fit

0:04:31.880 --> 0:04:32.880
<v Speaker 2>into the next decade?

0:04:33.000 --> 0:04:36.680
<v Speaker 1>Yeah, well, AI, I think can be amplification intelligence, not

0:04:36.720 --> 0:04:40.520
<v Speaker 1>just artificial intelligence and that amplification the superpowers that we get,

0:04:40.680 --> 0:04:43.520
<v Speaker 1>and it's not just part of super agency. Is if

0:04:43.560 --> 0:04:46.480
<v Speaker 1>you hit a superpower, that helps me too. That's how

0:04:46.520 --> 0:04:49.000
<v Speaker 1>we have superagency together, as long as.

0:04:48.920 --> 0:04:51.160
<v Speaker 2>It's democratic and everybody has it or is that a

0:04:51.240 --> 0:04:53.159
<v Speaker 2>question for in a second phase?

0:04:53.360 --> 0:04:55.440
<v Speaker 1>Well, I think one of the good things about it,

0:04:55.480 --> 0:04:58.080
<v Speaker 1>And that's part of the reason why the first chapter

0:04:58.240 --> 0:04:59.840
<v Speaker 1>is about humanity enters.

0:04:59.600 --> 0:05:00.880
<v Speaker 3>The chat at chat GBT.

0:05:01.360 --> 0:05:04.440
<v Speaker 1>When we build technologies for hundreds of millions and billions

0:05:04.440 --> 0:05:08.599
<v Speaker 1>of people, that's that's broadly inclusive. So that your uber

0:05:08.680 --> 0:05:12.359
<v Speaker 1>driver has the same iPhone that Tim Cook has. That's

0:05:12.400 --> 0:05:14.400
<v Speaker 1>the kind of inclusion that we're targeting.

0:05:15.120 --> 0:05:19.520
<v Speaker 2>Read I mean, I guess evolution is not necessarily progress

0:05:19.800 --> 0:05:22.120
<v Speaker 2>full stop. So how do you make sure that this

0:05:22.520 --> 0:05:24.800
<v Speaker 2>means progress for the majority of humans?

0:05:25.040 --> 0:05:25.800
<v Speaker 3>Well, so I think.

0:05:27.279 --> 0:05:29.800
<v Speaker 1>Look, I think as we iterate and we participate, we

0:05:29.839 --> 0:05:32.720
<v Speaker 1>make progress. And for example, even though you say, well,

0:05:32.760 --> 0:05:34.279
<v Speaker 1>we have a whole bunch of cars and that creates

0:05:34.279 --> 0:05:39.039
<v Speaker 1>climate change, the cars also create our industrial society. And

0:05:39.080 --> 0:05:41.000
<v Speaker 1>by the way, the way that we tackle climate change

0:05:41.000 --> 0:05:43.880
<v Speaker 1>is we add carburetor emissions and we add you know,

0:05:44.160 --> 0:05:47.760
<v Speaker 1>we new kinds of clean energy and we do evs

0:05:48.200 --> 0:05:50.920
<v Speaker 1>and so you know, I tend to be very you know,

0:05:50.960 --> 0:05:54.520
<v Speaker 1>as we do iterative deployment and as we bring humanity

0:05:54.560 --> 0:05:57.400
<v Speaker 1>into the loop, I tend to think we do make progress.

0:05:57.560 --> 0:06:00.320
<v Speaker 1>Now I think again, you make better progress by having

0:06:00.400 --> 0:06:04.239
<v Speaker 1>the right kind of design principles, by accepting criticism.

0:06:03.920 --> 0:06:05.320
<v Speaker 3>By talking about it.

0:06:05.520 --> 0:06:07.719
<v Speaker 1>So, you know, I describe myself as a bloomer, which

0:06:07.760 --> 0:06:11.680
<v Speaker 1>is not that technology is just great. It's technology engaging

0:06:11.720 --> 0:06:12.600
<v Speaker 1>with people as great.

0:06:12.839 --> 0:06:15.000
<v Speaker 2>But it also depends on the people in charge. What

0:06:15.040 --> 0:06:18.200
<v Speaker 2>do you think of Sam Altman's performance so far in

0:06:18.279 --> 0:06:19.080
<v Speaker 2>leading open AI?

0:06:19.440 --> 0:06:22.839
<v Speaker 1>Well, so I think, Look, I think Sam's great contribution

0:06:23.640 --> 0:06:26.960
<v Speaker 1>to humanity will be open AI. And that's what having

0:06:26.960 --> 0:06:29.600
<v Speaker 1>done a number of amazing things before and done amazing

0:06:29.640 --> 0:06:32.200
<v Speaker 1>investments like Confusion and all the rest. And I think

0:06:32.240 --> 0:06:36.240
<v Speaker 1>that his ability to think very big and to have

0:06:36.400 --> 0:06:38.960
<v Speaker 1>bet very hard on this, you know a little bit

0:06:38.960 --> 0:06:42.919
<v Speaker 1>of technological thesis of scale compute and scale learning systems

0:06:43.000 --> 0:06:46.039
<v Speaker 1>is what matters. And that's why open ai has brought

0:06:46.560 --> 0:06:51.440
<v Speaker 1>this current revolution to us. And it's these machines learn

0:06:51.880 --> 0:06:54.240
<v Speaker 1>and they learn things that we help them learn and

0:06:54.279 --> 0:06:54.960
<v Speaker 1>help teach them.

0:06:55.760 --> 0:06:57.920
<v Speaker 2>Is there someone I know You've also had your differences

0:06:57.960 --> 0:07:01.120
<v Speaker 2>with Elon Musk. Who's the person in the space that you,

0:07:01.440 --> 0:07:04.360
<v Speaker 2>if not admire or listen to the most well.

0:07:05.320 --> 0:07:08.680
<v Speaker 1>Sam Altman is definitely one of them. Kevin Scott at

0:07:08.760 --> 0:07:14.280
<v Speaker 1>Microsoft is another, Dario Ahmadi ananthropic as another, James Miyika

0:07:14.360 --> 0:07:16.800
<v Speaker 1>at Google as another. I mean, I think part of

0:07:16.800 --> 0:07:19.800
<v Speaker 1>the thing that's very important about making AI for humanity

0:07:20.240 --> 0:07:22.920
<v Speaker 1>is people who listen to others and talk to others

0:07:22.920 --> 0:07:25.160
<v Speaker 1>and accept criticism. And I think that's one of the

0:07:25.160 --> 0:07:27.400
<v Speaker 1>things that all of these people are very good at.

0:07:27.800 --> 0:07:30.360
<v Speaker 2>Do you need to regulate it or is it something

0:07:30.400 --> 0:07:32.960
<v Speaker 2>that you need to see how it runs and then

0:07:33.120 --> 0:07:34.720
<v Speaker 2>think about regulating afterwards.

0:07:34.800 --> 0:07:36.200
<v Speaker 1>So I think what you do is you start with

0:07:36.240 --> 0:07:39.600
<v Speaker 1>the absolute minimum regulation you could do for the things

0:07:39.600 --> 0:07:42.680
<v Speaker 1>that could be really bad, not for oh, look it

0:07:42.760 --> 0:07:45.080
<v Speaker 1>might have a biased picture or might have a bias statement,

0:07:45.160 --> 0:07:47.440
<v Speaker 1>like we can iterate, we can fix those as we're going.

0:07:47.960 --> 0:07:50.800
<v Speaker 2>Really bad is what people taking over planes to crash,

0:07:50.840 --> 0:07:51.720
<v Speaker 2>the things.

0:07:51.480 --> 0:07:56.240
<v Speaker 1>Like that, you know, cybercrime, et cetera. Regulate for that

0:07:56.960 --> 0:07:59.440
<v Speaker 1>right and then do iterative deployment. And by the way,

0:07:59.440 --> 0:08:02.440
<v Speaker 1>even though they deployment, you eventually get to fear the regulations. So,

0:08:02.440 --> 0:08:05.040
<v Speaker 1>for example, if you try to make everything perfect with

0:08:05.120 --> 0:08:07.880
<v Speaker 1>cars before you put them on the road, we'd never

0:08:07.920 --> 0:08:10.120
<v Speaker 1>have cars. So you put them on the road and

0:08:10.160 --> 0:08:13.280
<v Speaker 1>you go, oh, this for bumpers, this window wipers, this

0:08:13.360 --> 0:08:16.600
<v Speaker 1>for and occasionally, like the market doesn't want seat belts,

0:08:16.600 --> 0:08:18.520
<v Speaker 1>the car manufacturers all want seat belts, and then of

0:08:18.560 --> 0:08:20.600
<v Speaker 1>course the regulators going to say, oh no, no seat

0:08:20.600 --> 0:08:22.240
<v Speaker 1>belts is good, We're going to add those.

0:08:23.280 --> 0:08:26.360
<v Speaker 2>I mean, if you regulate for things, I mean terrorism

0:08:26.600 --> 0:08:29.720
<v Speaker 2>is bad actors, bad state actors, or so it's I mean,

0:08:29.720 --> 0:08:32.280
<v Speaker 2>how do you regulate it unless you have to protect yourself.

0:08:32.360 --> 0:08:34.920
<v Speaker 2>So it's basically finding the technology.

0:08:34.400 --> 0:08:37.080
<v Speaker 1>That blocks them. Well, I think it's the regulation is

0:08:37.120 --> 0:08:39.600
<v Speaker 1>if you're releasing the technology to the general public, which

0:08:39.640 --> 0:08:42.080
<v Speaker 1>could be to the bad actors as well, you're doing

0:08:42.120 --> 0:08:45.679
<v Speaker 1>red teaming and safety, you're putting the right security measures

0:08:45.679 --> 0:08:47.719
<v Speaker 1>to make sure that you're not bleeding the technology to

0:08:47.800 --> 0:08:51.080
<v Speaker 1>rogue states, terrorists, et cetera, that you have a safety

0:08:51.120 --> 0:08:54.120
<v Speaker 1>plan that if you go, Okay, why is the technology

0:08:54.120 --> 0:08:56.920
<v Speaker 1>I'm building if it does leak or anything else, why

0:08:56.960 --> 0:08:57.319
<v Speaker 1>will it.

0:08:57.320 --> 0:08:57.959
<v Speaker 3>Still be safe?

0:08:57.960 --> 0:09:00.719
<v Speaker 1>And how do we continue to have the technology that

0:09:01.440 --> 0:09:04.920
<v Speaker 1>makes anything that's in the wild as safe as possible.

0:09:05.280 --> 0:09:06.920
<v Speaker 2>On Nilon must Do you think he has too much

0:09:07.120 --> 0:09:09.000
<v Speaker 2>power being so close to the president.

0:09:09.200 --> 0:09:13.000
<v Speaker 1>Well, so, look, I think he's a celebrated entrepreneur.

0:09:13.160 --> 0:09:13.920
<v Speaker 3>But I think that.

0:09:15.400 --> 0:09:19.840
<v Speaker 1>Governments are not companies, like, for example, risking a company

0:09:19.960 --> 0:09:22.719
<v Speaker 1>like for example, to say, oh, our ten rockets blow up,

0:09:22.920 --> 0:09:26.640
<v Speaker 1>who cares, it doesn't matter the financial system of a

0:09:26.679 --> 0:09:30.560
<v Speaker 1>country blows up. That's the cultural revolution that's terrible. So

0:09:30.600 --> 0:09:33.000
<v Speaker 1>you actually have to say we take less risk here,

0:09:33.120 --> 0:09:35.920
<v Speaker 1>even at the price of some inefficiency, because it's more

0:09:35.920 --> 0:09:38.800
<v Speaker 1>important for us to not have things blow up.

0:09:39.280 --> 0:09:42.120
<v Speaker 2>Do you worry that things are going too quickly with

0:09:42.200 --> 0:09:43.920
<v Speaker 2>the Trump administration, Well, I.

0:09:43.880 --> 0:09:47.559
<v Speaker 1>Worry that very bad risks are being taken at speed.

0:09:47.679 --> 0:09:50.160
<v Speaker 3>Is not a problem. Risks are a problem.

0:09:50.480 --> 0:09:52.120
<v Speaker 1>And you know, for example, it's like, well, we're just

0:09:52.120 --> 0:09:54.000
<v Speaker 1>going to fire a whole bunch of people. Oh oops,

0:09:54.000 --> 0:09:56.600
<v Speaker 1>we fired a whole bunch of nuclear safety inspectors. Like

0:09:56.880 --> 0:09:59.960
<v Speaker 1>that's the kind of thing that's taking risks that is unwarranted.

0:10:00.559 --> 0:10:00.719
<v Speaker 3>Read.

0:10:00.760 --> 0:10:02.280
<v Speaker 2>I also want to talk to you about China because

0:10:02.320 --> 0:10:04.360
<v Speaker 2>Deep Sea kind of got everyone at the edge of

0:10:04.400 --> 0:10:08.000
<v Speaker 2>their seat. But where do you have a good understanding

0:10:08.040 --> 0:10:09.520
<v Speaker 2>about where China is on AI.

0:10:10.520 --> 0:10:11.960
<v Speaker 3>I have a reasonable understanding.

0:10:12.000 --> 0:10:14.760
<v Speaker 1>I do a fair amount of talking to various people

0:10:14.760 --> 0:10:17.920
<v Speaker 1>in China in order to make sure. Part of the

0:10:17.920 --> 0:10:19.560
<v Speaker 1>thing that when last year I was going around saying

0:10:19.600 --> 0:10:23.679
<v Speaker 1>there was actually an economic race with AI between the

0:10:23.679 --> 0:10:26.040
<v Speaker 1>West and China, people are like, oh, no, you're overblowing

0:10:26.040 --> 0:10:27.400
<v Speaker 1>that because you just simply don't want.

0:10:27.280 --> 0:10:27.920
<v Speaker 3>To be regulated.

0:10:28.160 --> 0:10:29.880
<v Speaker 1>And I think with deep Seak and everything else, we

0:10:29.920 --> 0:10:33.360
<v Speaker 1>see that that race is there, and the Chinese government

0:10:33.400 --> 0:10:35.800
<v Speaker 1>has said that they want to be you know, AI leaders,

0:10:35.880 --> 0:10:37.000
<v Speaker 1>like leading the world.

0:10:36.840 --> 0:10:39.080
<v Speaker 3>By twenty thirty. I think the race is on.

0:10:39.160 --> 0:10:41.960
<v Speaker 1>I think it's very important for US and our industries

0:10:42.320 --> 0:10:43.800
<v Speaker 1>to actually in fact be winning.

0:10:44.360 --> 0:10:46.360
<v Speaker 2>Is this an r Is this the new arms race?

0:10:46.760 --> 0:10:48.839
<v Speaker 1>Well, I don't call it an arms race because it's

0:10:48.840 --> 0:10:51.800
<v Speaker 1>primarily an economic race. There is arm stuff components with it,

0:10:52.240 --> 0:10:54.560
<v Speaker 1>but yes, it is the economic race.

0:10:54.880 --> 0:10:56.280
<v Speaker 2>Read thank you so much for joining us. That was

0:10:56.280 --> 0:10:59.160
<v Speaker 2>Reid Hoffman LinkedIn co founder Gray Luck, partner and of

0:10:59.200 --> 0:11:03.280
<v Speaker 2>course author of Super Agency. It's a good book. It's

0:11:03.360 --> 0:11:05.160
<v Speaker 2>well written, and it's to the point