a16z Podcast - a16z Podcast: For Your Ears Only

Episode Date: May 18, 2017

When it comes to spycraft — or rather, “tradecraft,” as they say in the biz — what do the movies get right, and what do they get wrong? In this episode of the a16z Podcast, Michael Morell — former Deputy Director and twice-Acting Director of the CIA — talks all things tradecraft and tech with a16z partners Matt Spence and Hanne Tidnam. What it is that the CIA really does? Is it a) James Bond, b) Maxwell Smart, c) Jason Bourne, or d) none of the above? For starters, it’s not at all about predicting what will happen — it’s figuring out what you need to know now to make the right decisions, asking the right questions, and reducing uncertainty. But that’s a tall order when you’re in the Situation Room advising the President — because there’s no such thing as zero uncertainty. So what makes the difference between a good analyst and a great one? How does technology affect tradecraft? And where do human spies come in? This podcast was recorded as part of our (now-annual!) podcast road trip, in conjunction with the a16z Tech Policy Summit, in Washington, D.C.

Hi, welcome to the A16Z podcast. I'm Hannah and I'm here today with A16Z policy team partner Matt Spence in Washington, D.C., to talk with Michael Morel, former deputy director and two-time acting director of the CIA about all things spies. We're here to talk about the art of spycraft. And first of all, is that even an okay word to use? Is that the right word? The word we use at CIA is tradecraft. What does that mean? What does intelligence gathering actually mean? So there's three different missions at the Central Intelligence Agency. One mission is to collect information from our adversaries. So think of it as stealing secrets. We recruit other human beings to be spies. We use technology to acquire that information as well. The second mission is then making sense of those secrets. That's what we call all source analysis and telling the president, here's how we think about this issue. A lot of people think we're predicting the future. That's not what we do. Oh, interesting. We're much more focused on describing the current situation. Why are we here? And what are the factors that are going to determine which way it goes?

That's what I spent my career doing, analysis. And then the third mission is covert action, which is somewhere between diplomacy and military action. Like we provided significant support to the solidarity movement in Poland, all the way to paramilitary activities like providing arms to the freedom fighters in Afghanistan when they were, when they were fighting the Soviet Union. That's covert action. It's kept secret. It can only be done at the direction of the President of the United States, actually a written document that he has to sign that says, here, I want you to do this, and here's the policy objective I'm after. Increasingly, as you think about that third bucket, intelligence is used in ways that sometimes blurs a line between what traditionally military

used to do. That's a lot by technologies. And sometimes a covert operation becomes not covert, you know, and involves the military. Some of that, especially the stealing secrets and the covert action sounds like some of the best spy movies. How is it different from what the public thinks of spying as, you know, when you're looking at Mission Impossible or what, what are the most common misconceptions? What do this public get right and what does Hollywood get right? And what do they get wrong? There are nothing but misconceptions. So nothing right? In Hollywood. There's a variety have different perceptions, and each of them is wrong. So one perception is the James Bond mission impossible perception is that there's not a secret we can't steal, there's not a plot

we can't stop. Superhero. Superheroes, right? Not true. This is a really tough business. There's a lot of secrets in the world that we need that we don't have, and so we're not perfect. Far from it. The James Bond myth is wrong. The second myth is that we screw up everything we touch. It's the Maxwell Smart, Get Smart myth, which is that we can do anything right. That myth is wrong. There's a lot of secrets we do steal. And then the third myth is the Jason Bourne myth, which is that there's a lot that the CIA does that's rogue that nobody in Washington knows about. That's a myth that could be at the end of the day most damaging because it paints a picture of a rogue secret agency operating in a democracy. People ask me all the time, is homeland real?

is a zero dark 30 real. Matt's smiling. And I actually laugh a little bit when I get asked that question. Here's what I say. I say not even close except for one thing. And that's the passion of the people doing the work. That is real. Particularly people who are working on life and death issues like terrorism, like the proliferation of weapons of mass destruction, human trafficking, drug trafficking on those life and death issues. The officers at CIA are just as passionate as the lead characters in Homeland and Zero Dark 30. I think that's right. I mean, the thing I would add to it based on, you know, spending time in the situation room is how uncertain the enterprise is. There is a sense, I think, that the public will have is that there is an answer. And if you just ask the right

person, you'll get it. And it is hard to determine the intentions of a foreign leader. So it's that. And the role, I think, intelligence plays when you make policy decisions. So you will not go to the CIA director and say, should we execute the operation against Assad bin Laden? Should we add 30,000 more troops to Afghanistan? Instead, you make a set spin of judgments about here's how confident we are. The fundamental job of intelligence is to reduce the uncertainty that the president is operating in. That is the purpose of the whole enterprise, but you can't reduce it to zero. Matt mentioned the bin Laden operation, and that is a great example.

We did not know for sure that he was in that compound. In fact, there was only a circumstantial case that he was there. There was no direct evidence. There was still a considerable amount of uncertainty that we left the president with about whether he was there or not. More than usual or less or about the same. About the same, I'd say. I remember actually a discussion in the sit room where people had different probabilities

of whether he was there or not. So the lead analyst said 95% chance he's there. And the senior most analysts had 80% chance. Wow. And I was at 60%. Wow, there was such discrepancy. And the president said to Leon Panetta, then the director of the CIA, who was sitting at the table. I was sitting against the wall because I was the deputy.

The president said to Leon Panetta, why is there such a gap here in these probabilities? And Leon turned to me and said, Michael, can you answer that question? And so I said, Mr. President, it's not because people are looking at different sets of information. you know, you can be assured that everybody has all the same information here. We're comparing apples and apples. We're comparing apples and apples. I said, I think that people are washing this data through their own personal experiences. So these analysts who came to work on terrorism post-9-11, who have known nothing but success

are washing this through their, that perception of success. I said, I'm washing it through the perception of the fact that we got the judgment wrong about Saddam Hussein having weapons of mass destruction. I was in that meeting, I remember. That statement, I think, was one of the most powerful statements I'd heard in hundreds of hours in situation room. I mean, to imagine that the president of United States is taking his presidency, asking you, how certain you. And that takes a huge amount of courage as an intelligence analyst to say, I remember those words. It was the meeting before the, it was the last meeting with the cabinet, before the president would walk down the colonnade from the Oval Office down to the residents, decide,

Is he going to launch the operation of Saddam bin Laden? And that takes courage to talk about the sheer uncertainty, especially compared to it. Well, in an enormous amount of self-awareness. And the punchline, right? The punch line was, Mr. President, the case that Saddam had weapons of mass destruction was stronger than the case that bin Laden is about-a-bott-a-bott. You could hear a pin drop in the situation room, and it was true, right? And that's just a great example of we reduce the uncertainty, but we can't get rid of it.

I was taking notes in this meeting. Michael said, I was the presence PDB briefer on 9-11. What is PDP? The presence daily intelligence briefing. Okay. So you're the person who tells the present about the major threats

and opportunities around the world and to have that perspective and Saddam Hussein Web was a master destruction and then present the information clearly, cleanly, not to try to tip the scales, but flaws and uncertainty all. There is a very sharp line between intelligence and policy.

So intelligence officers do not advocate policy, nor when they do their analysis or their intelligence reporting, should they have a preferred policy outcome in mind. Which, again, is a common misconception if you're looking at the Hollywood movie. Exactly, right? Why? Because you don't want a preferred policy outcome to influence what you collect or to influence your analysis or to influence how people see you when you're putting that intelligence on the, table, right? If I walked into the sit room and people knew that Michael thought that absolutely we should keep 20,000 troops in Afghanistan, then people are going to question the analysis I put on the table. But it's very interesting because part of what you were saying earlier in that

story of being aware of the filters through which we wash our own analysis, it takes an incredible awareness to think we are all coming to things with a particular set of experiences and we process stories in certain ways and come away with it. The way you talk about the analysis and the description using this information to set to describe things the way they are and then kind of parsing out all this, it's a kind of narrative, right? That you're first, you're piecing together a narrative for yourself

and then you're translating that narrative into something that you can tell to other policy officials to the president. How do you think about storytelling like that when you're gathering these facts to put them into one cohesive narrative? So the key is to understand your own biases because we try to be as objective as we can be, but we still have our own biases, right? We're still looking at the world through a particular

prism. And the best analysts understand their own biases. Can you do anything about it, though? Sure. We all make assumptions when we think through problems. One of the things that a good analyst does is say, okay, I think I just made five assumptions. Let me question each one of those. One of the ones that's on my mind right now. To make explicit and then to interrogate. one of the ones that's on my mind right now is this issue of can we count on China to squeeze the North Koreans and change North Korean behavior. There's an assumption there, which is that if the Chinese did squeeze them, the North Koreans would change their behavior. That's a huge assumption, and it's an assumption that has to be discussed and challenged and talked about, because if the

Chinese can't effectively squeeze them, then we shouldn't be asking the Chinese to do so. So these assumptions are embedded in every conversation we have, in every decision you make, and the more you examine them, the better off you are. But, Hannes, you said, but there's an artist storytelling, too. On the one hand, you need to be clear about your uncertainty and your assumptions. But if the president asks you questions, says, Michael, what will happen with North Korea and China? If you say, well, Mr. President, hear the seven assumptions, and you caveat your answer so completely, he'll just tune out and just not listen.

So the real pressure that people realize is there is an enormous problem. pressure to give an answer. You know, you're dealing with officials. I want to come in and say, what's the answer? Let's cut to the bottom line. And you need to find a way to give a bottom line with not making it so wishy-washy and hedged as to give them no answer. Because the danger is, then they just won't listen to you. And they'll listen to someone, but not the U.S. intelligence agencies and communities, which is equipped to provide them. Because it's a compelling story. Because it's a... Yeah, you need a story. I mean, the deal is there is so much information that comes at these decision makers. If you look back, again, to the

bin Laden operation. What was happening from January 2011 to May 2011 when the operation was planned? The president of China was visiting the White House, one of the most important geopolitical relationships we have. The Arab Spring was beginning. Military operations were beginning in Libya. There was the fall of the president of Egypt. And at the same time, the president was taking a trip to Latin America. I mean, with all that going on, you need some clarity, not because you want to take shortcuts, because the world is a messy place. You've got to make hard decisions. If you can't frame that into a package or a story that resonates with people who aren't intelligence experts, then you're just less effective. So the storytelling is really

important, and I think there's an art to it. And I don't think that most, I don't think we teach this to our analysts as effectively as we should. Oh, fascinating. What I always found is that the style in which analysts present their analysis is here's the conclusion, and then here's how we got there. And I don't think that's an effective way of telling the story. I think a much more effective way of telling the story is to say, here is the question that we were trying to answer. Here's how we went about answering it. And here's the answer we got. And here are the caveats to it. Which sounds like a classical essay structure, actually. It's much, much more effective. Interesting. The ability to answer a president's question without giving 10 caveats,

which no president is going to listen to is the difference between a junior analyst and a senior analyst, right? The junior analyst wants to give every caveat. You want them to. But one of the differences between junior analysts and senior analysts is people who have learned to move beyond that and have conversations with senior officials where they can be effective in providing information without going through every step of the process. It's interesting that in this job, sometimes the greatest mark of success is nothing happening, right? Or stopping something happening. And people don't know what they don't know. What's it like? to be in a job where your successes are kind of invisible. Your successes are largely invisible and your failures are on the front page. And so you have to develop a pretty thick skin because not only are those failures on the front page, but they're typically not 100% correct in how it's being told. Yeah, that must be really difficult to see a different version of the story. It's not that you can't fight back. It's hard to fight back because you can't put more classified information

in the public domain. So it's not easy. get now, if we can shift gears to talk a little bit about the mechanisms of tradecraft and how technology is impacting those, how it's changing. How is technology behind tradecraft, as I will appropriately call it from now on, changed not only how the job is done, but how you think of the job? So broadly, any advance in technology is both an opportunity and a problem, right? the opportunity is that you can use technology to do your job more effectively. On the collection side, cyber is a huge opportunity because it allows us to get inside other people's networks and to steal stuff. On the analytic side, technology, particularly big data and

artificial intelligence, gives analysts the ability to deal with much larger mass of data and to allow the analysts to focus on what's really important. Can you give some examples of how that's playing out. It plays out every day on the terrorism front where software is going through massive amounts of data, you know, going through a haystack to try to find the needle. And it doesn't pull out the needle, but it gives you a much smaller piece of the haystack to go through to find the needle. So that's the opportunity side. The cost side is that technology also gives our adversaries opportunities. So advances in encryption are used by terrorists to communicate with each other. Biometrics. Biometrics used at borders makes it very

difficult for CIA operations officers to travel in alias. So it cuts both ways. And it's a constant. So the technology is always evolving. As you guys know better than anybody, it's accelerating. And so there's always inside CIA, how do we take advantage of this and how do we defend against it at the same time? And the other thing that I saw in the seven years that I was in the White House and Defense Department is how much more important open source information is. Anyone can read it. But there's a mosaic that can be put together that technology allows. I remember several times that I would look through. And if you actually just look at what is not classified, you're getting as much information about certain types of topics than you could

otherwise. That would not happen without technology. I mean, even when I started in the government, you know, you would get a printout of news clips of readouts from the foreign press. So if you wanted to understand about Russia, you would read a translated version of some Russian newspapers. By the end, there was much more accurate, faster, deliver information where you could hear what was happening from the outside. So in this changing game of kind of information warfare through asymmetric information or bots and leagues, is it more of the same? Has there been a fundamental shift in how you think about things like misinformation?

During the Cold War, the Russians tried to interfere in U.S. elections, but the Internet wasn't available. And so, you know, it was one-offs. It was bribing a reporter to write a particular story, or it was placing a story in a particular place. You know, could have been fake news then as well, or true news, right? But what the Internet and cyber now allow is steal information on a massive scale, and it allows you to put it out in a massive scale. So within five minutes of Secretary Clinton falling in New York City on September 11th during the campaign, there were hundreds of thousands of social media postings coming out of Eastern Europe and Russia, questioning whether she was healthy enough to be president of the United States. It's the sheer magnitude of what technology and speed, of what technology and speed allow you to do.

It's the same game with a much more powerful tool. Interesting. And in this new world of this sort of quick and massive. massively scaled information and all the different new ways we have of gathering that information and image technology and surveillance. Is there a role for the human asset or is that going away? Is it just becoming a kind of large information game? It's a really great question and it's not a new question. So throughout the history of my 33 years at the agency, it was a question that was asked frequently about technology. Did our ability to collect telephone calls

and emails, and did that put the human spy out of business? Or satellite photography, right? Did that put the human spy out of business? And the answer is, is really twofold. One is each different type of intelligence collection has its strengths and weaknesses. One of the great, great strengths of human intelligence is that you can ask a question. So if you and I are having a phone conversation and that phone conversation gets intercepted. There's no way for the intelligence agency doing that to ask a question. There's no way to say, hey, but what about this and what about that? That's what a human agent can do. An unpredictable question or a question that we really need to know the answer to. A human spy can always ask a question, and that's really powerful. Technology actually sometimes

makes a human spy more important. So take encryption, for example. Terrorists can now talk to each other without security services and intelligence services getting in the middle of that. So if that's the case, then technical collection becomes not the tool of choice. What becomes the tool of choice is putting a spy at either end of those two points to know what's going on. So a human spy actually becomes more important as technology protects the adversary. What are some of the more creative ways of communicating information like that that you've seen? I mean, do you guys use the cloud? Do you use USB sticks?

Do you use paper? What are some of the more secure, creative ways of communicating that you'll – can you talk about? So this is always a huge issue inside CIA, right? Because the millennials want what they have on the outside. They want Wi-Fi inside CIA headquarters. Well, guess what? The counterintelligence people say, no way, no how, right? There are always these kinds of debates.

So for our most sensitive secrets, for CIA's most sensitive information, we don't store it digitally. It's paper. We store it on paper, in a safe, in a room with a lock on it that only a handful of people can get access to. How about out in the field? Possibly the same thing in the field. There's a huge downside to not storing information digitally, right? It can't be shared quickly. You can't search it and, you know, et cetera, et cetera.

But the cyber threat is so serious that for our most sensitive stuff, we don't store it digitally. There's classified information which are on classified networks, but you cannot forward one from, you can't sort of have a mistake and accidentally send it to someone's classified email address or their Gmail address. Things are separated from each other. When you have technology that's very hard to use, people to get their jobs done inevitably do things that are less secure. I mean, right now, I sort of joke, it's like we only now at the defense of where I got a Commodore 64 computer in Windows 95. There are ways to actually integrate and do these things that we just aren't putting, spending our time on. So a great example is the hack of the security files at the Office of Personnel Management. The cybersecurity at OPM was woefully inadequate.

If I could have stolen the same information from China, I would have done it in a nanosecond and I wouldn't have had to ask anybody permission. We left ourselves incredibly vulnerable. Cyber security in large U.S. financial institutions is light years ahead of cybersecurity in the United States government. There's a difference between defensive protection and going on offense. Is it different considerations when there's human bodies involved or when it's just technologically based? Do you think about these things in a different ethical lens? Matt and I were involved in many discussions in the situation room about this question. And I think where President Obama came down is we don't.

want to do something to another country that sets a precedent that would send a signal, right, sets a precedent for people to do it to us. Because guess what? We live in a glasshouse. We live in the biggest glass house. We are the most vulnerable, right? So when you're talking about offensive cyber operations, that's the line you draw. We better be very careful in taking the initiative because we're going to send a signal to the rest of the world that it's okay to do so. there were a number of discussions where if we go down this path, where does it lead? And that's a major question. There's also a question of will it work? Sometimes the bang for the buck may not really be worth it. You may take an enormous risk of something that may be discovered. And you actually think, how much would that make a difference? The issues that reach the president's desk in the Oval Office or reach him in the situation room are by definition incredibly hard. If they were not incredibly hard, someone else would have made that decision. Exactly. Sometimes you need to make this. difficult decisions. There's a lot of the intelligence can do, but it's no magic bullet. There just isn't a magic bullet. You've got to be, got to be really careful because something that somebody

wants to do might not be effective at all. So a great example. How are you, how do you measure that in advance that? Let me give you a great example. So ISIS is throwing out all of this propaganda, which is radicalizing young men and young women in Western Europe and the United States and Australia in other parts of the world. And so people say, well, why don't you just, why don't you just use your offensive cyber capabilities to destroy those servers, right, and stop them from doing that. And the answer is, because they can have another server up in an hour. You're using all these resources for very little gain, and you're setting a precedent that it's okay to use offensive cyber operations.

There is this tendency in every administration that I've seen, and I saw six of them, three Republican, three Democrat, for when the problem gets really hard to turn to CIA and say, fix this. Using covert action, fix this. And a lot of the times, CIA can't fix it. But CIA is, I don't know what the right word is. CIA is a pleaser. And so the CIA is very quick to say, of course we can do that. Of course we can do that. When actually, if you sit down and really think about it, no, you really can't. Can you give an example? Oh, I don't know if I can do that in an unclassified. That just shows how cool history in an unclassified setting. One thing I wanted to ask is how do you think about government oversight and executive branch action in situations where like every reasonable person would

agree secrecy is very important. How do you think about that kind of communication between agency and public? It's very hard. We have at our disposal a powerful intelligence collection network, a powerful opportunity for COVID operation at as limits, but is very powerful. But to set limits and say we are actually not going to do this, even though we can, is very, very hard. On the other side, there is not one piece of information. There will not be someone who's written a letter outlining their plans like the end of some movie where you pull off the mask and the villainism mask. It is a mosaic of a number of things that put together. The reason why law enforcement officers are trying to get more information, it's not that they're trying to take a

shortcut. You don't know what will matter. You don't know what individual piece will add up. Even again going back to the bin Laden operation we talked about you know there was no single thing that i think led this to go it is a cascade of information and you know what is individually valuable i would just add a couple of points one is we are not just trying to prevent terrorism and prevent the spread of weapons of mass destruction in order to protect people's lives we're actually doing it to protect our way of life some people are fond of saying that more people die in auto accidents or slips in the bathtub than terrorist attacks, questioning whether the government should be spending all these resources.

Well, the fact of the matter is, auto accidents and slips in bathtubs do not lead people to be willing to give up their constitutional rights. But a large terrorist attack, they are ready to give up their privacy and similarities in order to be protected. Right. But that's a big pendulum that swings back and forth.

And it does swing back and forth. But by point is that what we're doing on the security side is actually designed to protect. our privacy and our civil liberties. So that's that's point number one. Point number two, point number two is that there is a trade-off. Absolutely there's a trade-off and we have to realize it and we have to discuss it as a country and end up where we feel comfortable, right? And at the end of the day, it's where the American people want to be, educated by their leaders. On the encryption debate, I was actually not where you would expect me to be. I was not on the side of

the FBI. I was on Apple's side. Because by forcing Apple to create a key for its keyless encryption, you were creating another huge security vulnerability. If you create a key, it means a bad guy can steal a key. My preferred response to what U.S. high-tech firms have done to protect information with encryption is to say to the intelligence community, that's now a given. You've got to figure out a way to go get the information. So this brings us right back to the importance of those human agents, right? Find another way to get the information you need because forcing American companies is actually going to make matters worse. Even if you are working in Nash security, the intelligence community is really a black box

and hard to understand. What is the type of question you should ask an analyst about what can be predictive? I think the biggest thing that policymakers and even Silicon Valley can really learn about the intelligence community is what are the right types of questions to ask. You would not go to your doctor and say, who should I marry? I mean, maybe you do. It's a pretty dumb thing to do. You know, but like, they're right questions and wrong questions. So the key to good analysis is asking the right question. It's the same thing. It's the same thing, right?

You don't get insightful answers without insightful questions. So here's an example. If you look out the window and you see lightning and you hear thunder and you see dark rain clouds and you say it's going to rain, there's no insight in that judgment. at all, right? Because any non-expert could have come to that conclusion. Is it going to rain or not is the wrong question? Much more interesting and insightful question is, how bad is the storm going to be? Is there a risk of a tornado? Is there a risk of damaging hail or damaging wind? Right? Those are questions that it takes an expert to answer. So asking the right question is critically important

in any field. That reminds me even of the kind of work we do when we're editing. My experience is that briefings, oral briefings for senior policymakers are much more insightful for them than our written product. And the reason that's the case is because in the written product, we're asking the question. In the case of an oral briefing, it's the senior policymaker who's asking the question. The most valuable hour of my day when I was heading up Middle East policy, the Defense Department, was not reading the intelligence report. and I had a thick binder of fascinating things, but just having my briefery there so I could read it and ask the questions that would follow up.

The best briefers I had, and some ones who needed some more work was if you would ask a question, they would suddenly say, actually, I would think of it instead this way, you know? Well, it gives you the best results. Exactly. You may get a answer.

It may even be the right answer. But if it's the wrong question, you are devoting a lot of time and resource to something that is just going to send you down the wrong path. So one of the things that analysts do at CIA is they actually brainstorm. They have brainstorming sessions about what question to ask. Oh, like a newsroom. That can be used by any organization. I actually take some time and think about the questions you're trying to answer. So on the one hand, we have this sensibility

of this information gathering as being at the sort of leading edge of technological developments. But then you talk about government as being so, you know, as being very far behind where we would like, as compared to the private sector. So what is the intelligence community's relationship to current technology? So it used to be true until 30 years ago that the government was the technology

developer and the technology leader. So places, you know, CIA, we developed our technical tradecraft inside and the same thing was happening at DoD. But it soon became obvious in the mid-1980s, I'd say, that there was no way

the government could keep up with the advances being made the private sector and that the government was missing out on huge opportunities. And so people looked for opportunities to be able to tap into that technology. So we at CIA created something called Incutel, which is essentially a not-for-profit VC that invest in companies. We've co-invested with them in several times. And they invest in technologies that meet two criteria. one is that are commercially viable and two that would be valuable to CIA or the intelligence community 70% of what they've invested in have given a product or a service back to the

intelligence community so that government tapping in to what's happening in the private sector is incredibly important and that relationship is important now the really sad thing is that Edward Snowden did serious damage to that relationship because the perceptions of the private sector covertly helping the government, whether those were true or not. But the perception that was created, right, of that significantly damaged some American companies in what they were doing overseas. So there was a loss of trust among some high-tech firms and the government, right? And that's a shame because the government so relies on the private sector for the technology they need. Our government operate at two speeds. We either can move incredibly fast and get

something done, sometimes with a little clumsy and you kind of break things, or you're just so slow moving. It's hard to do things. The government can't bring people in fast even who are hugely committed to serve. Like, we don't bring new technologies in very quickly. And that's not a partisan issue. It's true for Democrat and Republican presence. It's true for the executive branch, as well as the Congress. That actually is one of those things. I think uniquely, the intelligence community is one of the organizations that I saw that does that better than most. That finds way to kind of innovate and think creatively. I had an old boss who I would walk into his office and say, I have a problem. And he said, Michael, Michael, Michael. There are only opportunities.

Oh, that's great. So I think that's a great way to look at the world. Thank you both so much for joining us. You're welcome. It's good to be with you. Thanks, Hannah.