
すでに機能している企業をどう変革するか――Leidos CIO アレクサンドラ・ギュンター氏に聞く
How to Transform an Enterprise That Already Works - with Alexandra Guenther of Leidos

すでに機能している企業をどう変革するか――Leidos CIO アレクサンドラ・ギュンター氏に聞く
The AI in Business Podcast
要約
Leidos の SVP 兼 CIO であるアレクサンドラ・ギュンター氏が、好調な組織ほど変革が難しい理由と、危機を待たずに変革の機会を見いだす方法を語った回。現状と将来の機会の間にある「ネガティブスペース」を探し、売上・利益率・リスク・従業員キャパシティという4つの価値レバーで変革の根拠を示すことを勧める。さらに、現場発のイノベーションの掘り起こしや変化への心理的抵抗への対処、エージェント型AI開発では「AIがすべきでないこと」を定義する発想転換について述べている。
この回のポイント
- ●ギュンター氏は、最適化され順調に動く組織ほど変革が難しいと述べ、現状の好調さの裏にある「機会の余白」をネガティブスペースと呼んでいる。
- ●変革の根拠は抽象的なイノベーション論ではなく、売上・利益率・リスク・従業員キャパシティの4つの価値レバーで示すべきだという。
- ●約5万人の従業員を抱える Leidos では、ネットワーク監視などの既存の観測データから現場の技術活用事例を掘り起こし、他部門へ横展開する取り組みを進めていると説明した。
- ●変化は闘争・逃走・硬直といった生理的反応を引き起こすため、リーダーは配慮したうえで明確で大胆なビジョンを示し、道筋は柔軟にすべきだと述べている。
- ●具体例として、人手を介さない請求処理(touchless invoicing)と、入社初日から顧客対応可能にする Day One readiness という大胆な目標を挙げた。
- ●エージェント型AI開発では、システムが何をすべきかより、何をしてはならないかというガードレールの設計に時間を割くべきだという見方を示した。
章立て
好調な組織ほど変革が難しい理由
最適化された組織では、機会の余白である「ネガティブスペース」が見えにくいと説明。Living systems thinking や「昨日の制約は今日の考慮事項では」という問いを紹介する。
ビジョンを現場に伝える:価値レバーと what-if
売上・利益率・リスク・従業員キャパシティの4レバーで成果を語り、what-if の問いで既存の制約から自由になる方法を述べる。
通常業務と将来構築の両立、現場知の活用
短期と長期の目標を併走させる考え方と、観測データから従業員の優れた技術活用を見つけ出し、組織内に広げる取り組みを紹介する。
変化への心理的抵抗への対処
変化が引き起こす闘争・逃走反応への対応として、まず配慮を示し、明確なビジョンで主体的な行動を促すことを語る。
大胆な目標の設定と具体例
1つの価値レバーに絞り、請求の自動化や Day One readiness といった大胆な一文の目標でクロスファンクショナルなチームを作った事例を話す。
リーダーへの実践的助言
エージェント開発で「してはならないこと」を設計する発想、組織設計への応用、節目の成果を祝うことの3点をまとめる。
解説記事
Leidos の SVP 兼 CIO、アレクサンドラ・ギュンター氏が登場した今回の The AI in Business Podcast は、「すでにうまくいっている企業をどう変革するか」がテーマ。業績が好調で顧客も満足し、危機感がない組織こそ変革が難しいという前提から、同氏が実践する考え方が語られた。なお、発言は本人の見解であり、Leidos やその経営陣の見解ではないと断り書きがある。
好調な組織に潜む「ネガティブスペース」
ギュンター氏によれば、成功している組織は数字、プロセス、役割、インセンティブ、ガバナンスが相互に結びつき、最適化されているため変えにくい。特に Fortune 500 のような大企業では、そこに複雑性も加わる。
同氏が重視するのは、うまくいっている点ではなく「何が可能か」「何を取りこぼしているか」という問いだ。うまくいっている部分の外側にある、さらに良くなる余地を「ネガティブスペース」と呼ぶ。人が技術を使って成果を出す仕組みとして組織を捉える「living systems thinking」という考え方も紹介された。CIO 就任時にチームへ投げかけた言葉は「昨日の制約は、今日の考慮事項ではないか」。固定要件とみなしていたものを手放す発想だ。
4つの価値レバーで変革の根拠を示す
変革を現場まで浸透させるには、大胆なビジョンを明確に伝えることが重要だと同氏は述べる。そのとき成果を語る共通言語が「価値レバー」で、営利企業なら売上、利益率、リスク、従業員キャパシティの4つになる。どのレバーを動かしたいかを示せば、従業員は自分の貢献が変革にどうつながるかを理解しやすい。
あわせて「what-if」の問いが使われる。このワークフローを守る必要がなかったら、昨日は経済的に成立しなかったことがAIで成立するなら、既存の機能ではなくキャパシティの出力を軸に組織を再編したら、といった具合だ。目的地は明確に保ちつつ、道筋には柔軟性を持たせるのが基本姿勢である。
現場の知を掘り起こす
約5万人が働く Leidos では、全員の工夫を把握するのは非常に難しいという。そこで、サイバーセキュリティや利用状況の把握のためにすでにある、ネットワークの可観測性・モニタリングのデータを活用し、従業員が技術を面白い使い方をしている例を見つけ出す取り組みを進めている。ユーザーは「企業内で最も活用されていないセンサー」だというのが同氏の見立てだ。ある部門の事例を、同様の課題を抱える別部門に共有することで、従業員同士のネットワークや新たな能力の横展開につなげている。
変化への抵抗と、大胆な目標の置き方
変化は人間に闘争・逃走・硬直の反応を引き起こす、という研究に触れ、リーダーはまず配慮を示したうえで、考える脳が戻るように明確なビジョンを提示すべきだと語った。全員を個別に橋の向こうへ連れて行くことはできないが、道を作り、進むべき方向を照らすことはできる。
ビジョンが曖昧ではないかという懐疑に対しては、1つの施策であらゆることをやろうとせず、1つの価値レバーに絞る規律が必要だと答えた。例として、経営陣が掲げた大胆な一文の目標が2つ挙げられた。1つは、政府系契約企業としての会計要件を満たしつつ、人手を介さずに請求を行う「touchless invoicing」。もう1つは、入社初日からクライアント対応が可能になる「Day One readiness」である。現実には人間の確認や例外対応は残るが、野心的な目標だからこそ部門横断のチームを組みやすいという。
エージェント型AIでは「しないこと」を設計する
締めくくりの助言は3点。第一に、エージェント型開発における人の役割を見直すこと。従来のソフトウェア工学では、システムが何をどう処理すべきかの定義に開発時間の大半を使う。エージェント的な発想では、AIが何をすべきでないか、どんなガードレールや制約を置けば想定外の逸脱を防げるかに時間を割く必要があると述べた。「ネガティブスペースをコードに書く」という発想の転換である。
第二に、同じ考え方を組織設計にも適用すること。成果、意思決定権、譲れない条件、リスク、従業員の自律性を定め、従業員を将来の役割づくりに参加させる。第三に、短期・長期の節目を祝うこと。技術面では1年超先を見通すのは難しいとも話した。
まとめ
本エピソードの要点は、危機を待たずに変革の理由を自ら作るための道具立てが示されたことだ。4つの価値レバーや大胆な一文の目標は、日本企業でも、好調な事業での変革や部門横断のAI活用の合意形成に応用しやすい。また、可観測性データから現場の活用事例を拾う発想や、エージェント開発で禁止事項や制約の設計を重視する視点は、AIガバナンスを検討する技術リーダーにとっても参考になるだろう。
文字起こし(英語・自動生成)
Welcome, everyone, to the Emerge AI in Business podcast. Today's guest is Alexandra Gunther, SVP and Chief Information Officer at Lado. In today's show, Alexandra discusses why the hardest organizations to transform are often the ones that already work well, and how leaders can find the negative space between current performance and future opportunity instead of waiting for a crisis to force change. She explains how to build the case for that kind of change using concrete value levers like revenue, margin, risk, and employee capacity, how to service innovation already happening on the ground, and why authentic AI development should start by clearly defining what the technology should not do. Just a quick note for our audience, such as the views and opinions expressed by Alexandra on today's program are her own and do not reflect those of Lados or its leadership. getting in front of enterprise AI buyers
is not about impressions it's about trust at Emerge we help AI vendors engage decision makers through research driven content and conversations that actually matter in the buying process to learn the exact strategies we use to help leading AI brands and startups connect with their ideal enterprise AI buyers visit Emerge.com slash AD1 that's E-M-E-R-J dot com slash AD1 now the conversation with Alexandra Alexandra, welcome to the Emerge Podcast Studio. Thank you so much for having me. I'm super excited to get into conversation with you today. And I think that what we're about to speak on is not necessarily on everyone's radar. and that is what excites me. There's a strange kind of problem that only shows up in organizations that are doing absolutely fine.
Nobody's getting fired over last quarter's numbers. The clients are happy. Everything just looks great. There's no crisis forcing anyone's hands. So the picture looks perfect. And I want to get into this. When the organization isn't broken, what makes changing how it operates so difficult? Yes. Successful organizations can be the hardest to transform. Because they're optimized, like you said. They are running fine and they're making their numbers, their processes, roles, incentives, governance, all interconnected and all operating. So when we think about what makes change needed and change possible, that's where we start thinking about the idea of opportunity space. I call this the negative space because when we're looking at what's working, what we're missing is the opportunity space for what could be even better. And the other complexity we have when we look at organizations that are well-tuned, if they're large, especially, say, Fortune 500 enterprises like Leidos is, you have an interconnected set of complexities that adds another layer to this.
So we've got opportunities, and then we also have complexity that need to be considered. So as a technology leader, when I come into an organization and look at not just what's working, my biggest question is, what's possible? What are we missing out on? And I have an approach to this. I call it living systems thinking. So as a technologist and a systems engineer, I think about how everything's interconnected. did. As a leader, I think of how humans are operating the system and optimizing for outcome. And so this idea of living systems, humans using technology for amazing outcomes, this is where you start being able to architect for the negative space. So when we think about negative space, this idea of looking for opportunities, you can get there by asking different questions. And so one of the best phrases I like to introduce to my team when I became CIO is, what if yesterday's constraints are today's considerations?
You release this idea that it's a fixed requirement. And so as we look at what's possible for the business, for employee capacity, for markets, what can we reconfigure and rethink about in new and bold ways that will not just allow our employees to achieve different outcomes, but empowers them to think more boldly as well? Yeah, I like that. And I think, obviously, the type of person that thinks like this in an organization is normally at the top. It's someone in leadership and they will have to give this idea down to whoever is on the ground. So how does that work Giving the handoff of this is the idea this is the opportunity that we see to someone on the ground thinking but there nothing wrong with what we doing right now Yes this is where the idea of being a really strong bold visionary and being able to communicate that vision effectively for buy pays dividends
So you want to use a forward-back kind of philosophy here as a leader looking to garner that buy-in from the CEO down to your frontline workers to understand the value you're delivering. So if you need a strong, bold vision that will engender that kind of followship, if you will, then you need to be really articulate about what your business outcomes are. Now, when we talk business outcomes, I actually talk value levers because every business, whether you're for profit or not for profit, you have value orientation. So for-profit businesses, I use revenue, margin, risk, and employee capacity. And when we start talking about, so what does the future look like? What kind of bold things? Where do we want to move the needle on each or some of those four levers? That helps employees see their direct contributions to those changes.
And so you then want to use a what-if framework. What if we didn't need to preserve this workflow? What if an organization can provide something that was economically not feasible yesterday, but is feasible in this new age with AI, with different technological operating models? What if we reorganized around capacity output instead of the functions we already have, right? This idea of letting go of constraints also applies for things like organizational design, as well as operating models and technologies. So this forward-backed mindset combined with this release from what was previously constraints, it gives us flexibility on the route as we are crystal clear on the destination. And I think as a leader, that's the kind of approach you want to take to engender the most fellowship in your employees as well as your executive team and your external customers.
Also, as a leader, I think the challenge is also going to be running the business at full speed, knowing there's nothing really wrong with it. We might just be missing out on opportunities. How do you run it at full speed while you're also building the future of the organization? Yes. So, I mean, it requires a dual look at both near-term and long-term goals. And in order to keep followership and to keep achieving near-term goals, you need to have those clearly articulated. You also need to make sure you're making progress towards that destination, those larger value items that you are really targeting for a longer time period. One of the things I actually have been working internally with our teams about and was just posting about online was that our users may be the best untapped sensors that we have in an enterprise. And so our users are doing incredible things every day. Lighthouse is a large company. We have 50,000 people around the globe. Understanding what everyone is doing and the innovative concepts they're bringing forward and technologies that they are using to enable outcomes, it's, let's say, a tall order, right?
It's extremely, extremely difficult to source all those good ideas and good outcomes and achievements and be able to then elevate and scale across a broader team. And what we're using is using employee experience and technology in order to source opportunities where we are seeing employees use technology in interesting ways. And based on observability and monitoring that's happening across our networks anyway for cyber reasons and for usage reasons, et cetera. And we're actually tapping into that to say, what if we can mine use cases through that? And so that's a really great way for leaders to be able to mine information they already have. But again, this idea of like looking at it differently. How else can I use this information? What if I can source grassroots goodness and use that to say, oh, this other business sector has a similar problem and they have a similar team approaching it with their engineers.
Can we share this out? So it's been a great way to create networks across employees and also opportunities for them to source net new capabilities and then replicate it across the large organization. That sounds like a simple recipe to follow, but obviously we know everything cannot always work out the way the recipe wants it to. So is there a point where showing someone this bridge just isn't going to be enough where you would have to let them walk across it themselves? Yes, change management is real. And I love that with this advent of not just AI, but just the pace of change across the technology field, we been seeing a resurgence in organizational change management and change techniques come forward as a it not just a nice to have it now a must do And when you talk about how do you help people who are not walking across that bridge get there and how do you make sure you've got full fellowship that you're shooting for,
your organizational change management practices come into play. And I read some really interesting studies and information about the psychology of change. And so humans see change and it triggers a fight, flight, hide response. There's a biological, physiological response to change. And so when this is triggered, what your response is as a leader is to first give grace, obviously, to understand that this is happening and to really, really quickly make sure you pull forward that bold vision. You have to give them something that helps their thinking brain come back online and see alignment on what is possible to what they themselves can get on board with. And so a crisp, clear vision is really, really key. And then helping them get out of that spot and empowering them to take steps themselves. Because you can't walk every individual across the bridge individually, but you can create a path. and shine a spotlight that helps them focus on where they need to go.
So, again, protecting that destination, but allowing for flexibility in the path. Employees are going to have different paths to get there, and they're going to find some amazing things along the way, and sometimes there's going to be stumbles. And so, as a leader, our responsibility is to make sure the destination is crystal clear in a way that inspires both fellowship, accountability, all of the things that make it necessary for employees to get there. We know from experience that there will always be that one skeptic, maybe more than one, especially. Let me be honest. I'm the person that I don't like change. I go in fight and flight mode all at once when something needs to change. But there will always be this one thing, you know, we're calling it a vision, we're calling it an end destination, but no one really knows what it's going to look like. It's still very much imaginary. So if the underlying data and the systems aren't even connected yet, isn't showing someone the other side or the vision not really an honest picture? How do we manage that angle? Yeah. So there's a couple of ways to do it.
This is also why I built the value levers that I did for the latest teams to focus around, are we increasing revenue? Are we increasing margin? Are we looking to reduce risk? And or are we really focused on employee capacity? for this particular initiative. This is where discernment and discipline becomes a hallmark for good leader. You need to not try to do everything with every initiative because that's how you lose people. That's how you lose your skeptics. That's how you start to break the strong light you have on the path towards that destination. So one of the things that we did across the Leidos Leadership Team, We took a look at core cross-functional things we wanted to change about how we operate as a business in order to make ourselves more effective, more efficient, and more transparent with our customers. And then we created a bold statement around it.
So what I advocated for is give us one statement that is bold, it's a little bit audacious, and then it allows us to craft a strong scope and cross-functional teams around it. So one is touchless invoicing. So the finance team came forward and said, my bold goal would be that we can compile and aggregate all of our costs. And as a government contractor, there is a ton of accounting guidelines that go into that, making sure we are capturing everything audit ready, but then also billing appropriately and securely our customers as needed. What if we could do that without a human in the loop? Now, the reality is we all know a human will be in a loop and obviously reviewing things and exceptions. but you have to create like an initial, you know, they say shoot for the stars and you'll wind up among them. And so we want to make sure we're shooting for a bold goal. Another one is we are a people business. We have 50,000 people around the world. We want our employees to be effective on day one, right? And so we have this day one readiness goal. What happens if you were
fully client ready on day one? And in reality, you need your assets, your access, your information from your project manager and your client-side information all pulled together at the same time in order to be day one ready, in order to be client-side ready on day one. And that is a bold set of cross-functional items. And so creating these captured, bold statements allow us to then say, employee is like, can you get behind this? Would you want to be part of a group that's looking to transform how we're invoicing our customers? So it's looking to transform how every employee from here on forward comes onto the light of team. I love that. And I love that you take the different personalities into consideration in this plan. And definitely, can I say girl map when we say shoot for the stars, have that bold statement because then everything that not that bold it acceptable So a little bit of mind games going on here And you given our leaders a lot to work off already but to wrap things up and to give them some clear practical guidance
for our senior leaders listening, what should they focus on when moving their people from the old way of working to the new one? Yes, you want to design for both the guardrails for technology and the conditions needed for human adaptation. So especially as technology leaders, we want to continue that strength in strong technology controls and governance and guidelines, but also build for the human optimization of the system. And so that, in reality, looks like a couple of things. So one is rethink the human role in development. When we talk about agentic development in particular, this concept I want to bring forward, we talked earlier about this idea of negative space and opportunity space. It also applies for agentic development. So when we're looking at technology, leadership, and guidance, traditional software engineering principles really devote the vast majority of coding time
to developing the solution, articulating what the system needs to do, how it needs to do it, how it needs to process data, and what kind of output it needs. When you shift to agentic thinking, you need to flip the script and spend all that time now saying what the AI should not do. What guardrails do you put on it? What constraints do you put on the AI so that the system does not have one of these escape clauses and escape conditions happen? And so flipping the script on, you need to actually code for the negative space is a different way of thinking. So the first one is rethink the human role in agentic development. And then the second piece of the leader is to apply the same thinking to organizational design. So you're going to define outcomes. You're going to think about decision rights, non-negotiables, risk, autonomy of your employees, and then invite employees to participate in what that future set of responsibilities look like for them and make sure you're redesigning things like relationships and races
and articulating that as you go so people feel grounded in what they know and can still shift with the change. And then the third piece is remember to celebrate. Celebrate both your near-term and your long-term marker accomplishments. And so as you go through day in and week out and look at what's coming in the next month, the next quarter, the next year, going more than a year out technologically right now is pretty darn hard. It's requiring a lot of crystal ball. And so we want to make sure. Oh, go ahead. No, absolutely. That's 12 months. Thinking of 12 months, hey, we don't even know what's going to happen in three months. Yes. So this idea of rethinking what's required from our technologists within our organization, especially in agentic development, you bring that negative concept forward, making sure we're also applying that same concept to organizational design. And don't forget, we humans love to celebrate. and that is really necessary to maintain your fellowship of your employees and your teams
and articulation of that bold vision to keep them focused. I love that. I wanted to wrap up this by basically summarizing our conversation, but I don't even want to distract our listeners. I want them to really think about this last part and listen to what you had to say and the guidance, because I think it's very, very practical, but also extremely relevant. Alexandra, it's been an absolute pleasure having this conversation with you. Thank you so much for sharing so much of your top secrets. Wonderful. I am happy to come anytime. Thank you. Wrapping up today's episode, let's look at the three key takeaways from our conversation with Alexandra. First, the hardest organizations to transform are often the ones that already work. So leaders need to find the negative space between current performance and future opportunity rather than wait for a crisis.
Second, building the case for that change depends on concrete value levers. Revenue, margin, risk, and employee capacity all matter, not abstract appeals to innovation. And finally, as organizations adopt authentic AI, the more durable design question is defining what the technology should never do, not just what it should or could do. Getting in front of enterprise AI buyers is not about impressions. It's about trust. At Emerge, we help AI vendors engage decision makers through research-driven content and conversations that actually matter in the buying process. To learn the exact strategies we use to help leading AI brands and startups connect with their ideal enterprise AI buyers, visit emerge.com.ad1. That's E-M-E-R-J.com.ad1. For further executive-level analysis and to join our network of leaders delivering workflow impact with AI, visit Emerge.com. On behalf of the team at Emerge, we'll see you on the next episode.
番組の概要欄(原文)
The hardest enterprises to transform are often the ones already succeeding, where there's no burning platform to justify change. In this episode, Alexandra Guenther, SVP and Chief Information Officer at Leidos, explores how leaders find the negative space between current performance and future opportunity, and build a case for change using concrete value levers like revenue, margin, risk, and employee capacity. She also discusses how to surface grassroots innovation already happening inside the organization, manage the psychological resistance that accompanies change, and reframe agentic AI development around clearly defining what the technology should not do. Enterprise AI buyers want to know where your solution fits and what value it can deliver. See how Emerj helps AI vendors earn their trust through research-driven content at emerj.com/AD1
出典・この記事について
- 番組
- The AI in Business Podcast(Daniel Faggella)
- 元の回
- How to Transform an Enterprise That Already Works - with Alexandra Guenther of Leidos
- 配信日
- 要約作成日
- 公式サイト
- emerj.com
この要約・解説記事は、元の音声を文字起こしした内容をもとに AI(Claude)が日本語で作成したものです。 固有名詞や数値に誤りを含む場合があるため、正確な内容は元の音声でご確認ください。制作方法について
関連エピソード

ServiceNow投資からさらに運用価値を引き出すには――BigPanda CEO アサフ・レズニック氏が語る
The AI in Business Podcast

連合学習でエージェント型AIを強化する:Flower Labs の Lorenzo Sani 氏に聞く
Practical AI

10年使える計算プラットフォームの選び方 ― IDEMIA の Kyle Larson 氏に聞く
The AI in Business Podcast

AIがブランド価値を剥ぎ取る:Matt Clare(UJet)との対談
VUX World

AIモデル投資の前に決めるべきこと:「仕事」を定義してからモデルを選ぶ - Salesforce Melissa Ramey氏
The AI in Business Podcast

データとセマンティックレイヤーを整えて製薬AIエージェントを正解に導く――Flatiron Health リチャード・ノーキ氏
The AI in Business Podcast