• eople stopped contributing. He was the only one solving problems, which meant the department's capacity for improvement was limited to whatever he could personally see and direct.
• Your job as a leader is to improve performance, but the way you get there is not telling people what to do — it's engaging everyone every day and asking for their ideas.
• “I actually liked my job for the first time as a leader. Running around implementing people's ideas is fun. Running around telling people what to do is not.”
• This is the part I think a lot of organizations should pay attention to, because it's the most common failure pattern in continuous improvement. You train everyone. You believe in the principles. And then nothing fundamentally changes, because training gives people knowledge but not a system for applying it day after day.
• a set of 10 core processes covering strategic planning, annual goal setting, continuous daily improvement, large-scale improvement events, and standards of respectful behavior. Every year, the senior leaders sit down and ask how they want to change the standard processes. Every version gets a little better.
• They didn't design the perfect system and roll it out. They started, standardized, and improved — version after version. That's a fundamentally different approach than what most organizations do, which is design a big system, launch it, and hope it sticks.
• Level one is the executive space — a physical room with a North Wall showing the organization's True North (“Best place to give care, best place to get care”), performance metrics starting with patient-oriented measures and ending with financial ones, strategy maps, and accountability boards.
• Level two cascades to divisions and service lines — aligned with level one but adapted to their scope
• Level three is where the real action happens: the frontline idea boards.
• UMass Memorial now has 650 active teams running what they call their “digital innovation station” through KaiNexus
• Through this system, they've implemented 200,000 frontline staff ideas. When I first interviewed Eric in 2015, that number was around 2,200.
• But here's what I think that number really tells you. It's not just 200,000 good things that happened at UMass Memorial. It's 200,000 proof points that the people doing the work knew things that leadership didn't. Every one of those ideas existed in someone's head before the system gave them a way to surface it.

• “Nobody wants to implement MY idea. They want to implement THEIR idea, and they will work 10 times harder to make sure their idea is successful than they will for my idea.”

   

      
• This isn't just a philosophy of respect — though it is that. It's a prediction about execution. Ideas pushed down from the top tend to get implemented poorly, slowly, or not at all.
• If the people doing the work don't feel ownership, the acceptability drops and so does the follow-through
• He can pull up his phone and look across all 650 idea boards to see what teams are working on–using KaiNexus.

• Eric is describing unfiltered, real-time visibility into frontline activity across 150 sites. That's not just a nice-to-have. It's an intelligence advantage that most health system CEOs don't have.

      
• When other leaders say they can't find the time for that, Eric's response is pointed: of all the things you want to do, talking to the people closest to where value is created for your customers is the one thing you have to make time for.
• The bond rating went from junk status to an A
• Those results didn't come from a single initiative or a training program. They came from building and refining a management system, year after year, that connects frontline ideas to strategic goals
• But what struck me was how he's approaching AI implementation. He's insisting on A3 thinking first: don't implement an AI tool unless it's addressing the root cause of a real problem. Don't push solutions down onto the workforce because a vendor told you it was great or because another organization tried it.
• start by preparing the workforce, educate people on what it actually is, and then apply it within a disciplined problem-solving framework.

• “We don't need a human in the loop when it comes to AI. We need a human in the lead.”

   

      
• Most organizations are going to rush into AI the same way they rushed into Lean — training people, buying tools, pushing solutions.

22 more annotations...

• Fall—not fail—fast

   
• Sinek, however, said HR should avoid the “failing” language “at all costs. Instead, let employees feel comfortable “falling.”
• They can fall—a lot—pick themselves up and start running again,” he said.

• Data-informed—not data-driven

   
• many HR professionals are taking a “data-driven” approach to decision-making, but being too reliant on data may suggest a lack of long-term vision,
• “It’s important to be data-informed—not data-driven,”
• “When people abdicate responsibility—whther to a lawyer or to data—it shows a lack of vision,”
• You’re an amazing team’
• Leaders, while well-intentioned, often overuse the second person in trying to motivate their teams
• You did this.’ I find it a little condescending,” he said. That language tends to create separation between leaders and employees, and among employees themselves.
• Shift instead to first person: “‘Look what we did. I’m so proud to be part of our team.’”

9 more annotations...

• The CIO felt the answer was obvious: agentic AI systems roll up to her. The COO countered that an agentic workforce is the definition of ops. The CFO noted that an AI system was already making underwriting decisions with direct P&L impact. The Chief Risk Officer observed that autonomous decision-making systems are a major risk exposure. The CHRO saw AI agents as functionally equivalent to workers, which partly fall under his remit. The Chief Data Officer reminded everyone that the whole kit and caboodle depended on the right permissions and data access, and this was her domain.
• professional groups are locked in a perpetual contest over who controls which domains of work, and that major technological or social disruptions are the moments when those boundaries get redrawn.
• Abbott argued that the professions exist in an interdependent system, perpetually competing with one another for control over domains of work.
• Jurisdictional competition has long existed inside organizations as well.
• The arrival of ERP systems, for instance, created tension between business process owners and IT.
• As cloud spend ballooned and security risks and compatibility issues surfaced, the CIO reimposed central control over the shadow IT orgs that popped up in business units.
• Agentic AI will create far more radical change. Returning to the insurance company, an AI agent that autonomously processes claims truly is a technology system (CIO), an operational workflow (COO), a risk management concern (CRO), a financial decision-maker with P&L impact (CFO), a novel form of labor (CHRO), and a user of data (CDO).
• The CIO’s claim rests on infrastructure.
• The COO sees agentic AI as a component of the core ops of the firm.
• The CHRO’s claim is conceptually provocative. Agents are functionally a new category of workers. They need to be trained on company-specific knowledge, enculturated, supervised during deployment, evaluated on performance, and paired with human colleagues in ways that preserve morale, culture, and institutional knowledge
• The CFO’s claim rests on economic accountability.
• Risk and legal leaders see autonomous systems as a novel form of failure, liability, and regulatory exposure
• Finally, the CDO notes that every other functional claim is downstream of the data.

• From “Who Owns AI?” to “Who Owns Which Decisions?”

   

• organizations typically default to one of four responses:

   
   
  • Hand it to the most capable leader
   
  • Give it to the squeakiest wheel
   
  • Ship it off to a committee
   
  • Pass it to whoever has existing budget
• We need a different approach. That approach should begin by shifting the question from “who owns AI?” to “who owns which AI-related decisions?
• decomposes the problem into specific decisions and assigns each to the person best positioned to make it.
• To make the shift from ownership to decision rights, companies will need to put into place a decision and accountability map.
• This returns us to the Chief AI Officer. Many companies are creating this role out of confusion about how to proceed, because they recognize either that AI ownership can’t belong to a single individual, or because no one on the current executive team has the expertise or bandwidth to lead the agentic AI adoption process.
• What a CAIO can own is the coordination layer. In practical terms, this means owning the decision rights map itself: creating it, maintaining it, identifying gaps, and driving for clarity when assignments to functional heads are unclear.
• This is a genuinely novel jurisdiction. No incumbent executive has ever owned the cross-functional orchestration of autonomous systems
• None of these steps will eliminate jurisdictional competition, but they will channel it.
• It is important to underscore that there is no permanent answer to the question of who should own which AI decisions. The technology is too broad, too significant, and too fast-moving for any settlement to hold
• <div class="Standard_content__mghDk"><p class="Paragraph_text__yGFNO">It is to design the map that ties ownership to implementation, and then to revisit it every time the technology outpaces the current version of the org chart. Which, for the foreseeable future, will occur often.</p></div><div class="Standard_right-rail__tTaUI"><div class="Standard_right-rail-content__Y1Qwu"><div class="Standard_piano-container__3yB5_"><div data-testid="piano-ad-container" class="right-rail-piano-container advertising PianoAd_ad__m1O2b"><div class="tp-container-inner" style="height: 605px; width: 300px;"><iframe frameborder="0" allowtransparency="true" scrolling="no" name="offer_fb5c910417a633984366-0" id="offer_fb5c910417a633984366-0" style="overflow: hidden; background-color: transparent; border: 0px; width: 300px; height: 605px;" src="https://buy.tinypass.com/checkout/template/cacheableShow.html?aid=mbs77qtzUz&amp;templateId=OT7LSF73JJQ3&amp;offerId=fakeOfferId&amp;experienceId=EXJKATV2ZT1Z&amp;iframeId=offer_fb5c910417a633984366-0&amp;displayMode=inline&amp;debug=false&amp;pianoIdUrl=https%3A%2F%2Fid.tinypass.com%2Fid%2F&amp;widget=template&amp;url=https%3A%2F%2Fhbr.org%2F2026%2F03%2Fwho-in-the-c-suite-should-own-ai" allowfullscreen="true" allow="payment"></iframe></div></div></div><div class="Standard_right-rail-middle-container__xc6DH"><div data-testid="ad-container"><div data-google-query-id="CKvexOaBopMDFSdOCAQdlig8ow" data-testid="ad" class="" id="right-ad"><div id="google_ads_iframe_/34363400/HBR_300x600_1__container__" style="border: 0pt none;"><iframe marginheight="0" frameborder="0" scrolling="no" marginwidth="0" name="google_ads_iframe_/34363400/HBR_300x600_1" id="google_ads_iframe_/34363400/HBR_300x600_1" data-integralas-id-ad1d8e18-5654-aab0-7224-9a9320442320="" style="border: 0px; vertical-align: bottom;" height="600" title="3rd party ad content" data-google-container-id="a" data-load-complete="true" tabindex="0" allow="private-state-token-redemption;attribution-reporting" aria-label="Advertisement" width="300"></iframe></div></div></div><div class="Standard_recommend-right__MBjxq ecommerce-module"><div data-testid="recommended-for-you-container" class="content-rec-wrapper RecommendedForYou_container__icYLj RecommendedForYou_standard__tIiRH RecommendedForYou_right-rail__nSrKs"><div data-testid="container" data-analytics="placement-recommended-for-you-right-rail"></div><h4 class="spacing-narrow">Recommended For You</h4><ul class="at-rightrail"><br/> <li><br/> <div style="display: flex;"><br/> <figure class="product-thumb"><br/> <a rel="nofollow" href="https://hbr.org/2026/01/dont-underestimate-the-value-of-professional-friendships?ab=at_art_art_rr_v1x4_s01"><br/> <img alt="" src="https://hbr.org/resources/images/article_assets/2025/12/Dec25_19_2190551887-383x215.jpg"><br/> </a><br/> </figure><br/> <div class="hed title-ellipses-3 title-h4" js-target="ellipse-product-title"><br/> <div class="at-topic"><br/> Personal Growth And Transformation<br/> </div><br/> <a rel="nofollow" href="https://hbr.org/2026/01/dont-underestimate-the-value-of-professional-friendships?ab=at_art_art_rr_v1x4_s01">Don’t Underestimate the Value of Professional Friendships</a><br/> </div><br/> <stream-item data-image="https://hbr.org/resources/images/article_assets/2025/12/Dec25_19_2190551887-383x215.jpg" data-authors="Paul Ingram" data-content-type="Digital Article" class="stream-item" data-topic="" data-id="tag:blogs.harvardbusiness.org_2007-03-31:999.420668" data-url="https://hbr.org/2026/01/dont-underestimate-the-value-of-professional-friendships" data-title="Don’t Underestimate the Value of Professional Friendships"><br/> </stream-item><br/> </div><br/> </li><br/> <li><br/> <div style="display: flex;"><br/> <figure class="product-thumb"><br/> <a rel="nofollow" href="https://hbr.org/2026/02/ai-doesnt-reduce-work-it-intensifies-it?ab=at_art_art_rr_v1x4_s02"><br/> <img alt="" src="https://hbr.org/resources/images/article_assets/2026/02/Feb26_07_EynonJones-383x215.jpg"><br/> </a><br/> </figure><br/> <div class="hed title-ellipses-3 title-h4" js-target="ellipse-product-title"><br/> <div class="at-topic"><br/> AI And Machine Learning<br/> </div><br/> <a rel="nofollow" href="https://hbr.org/2026/02/ai-doesnt-reduce-work-it-intensifies-it?ab=at_art_art_rr_v1x4_s02">AI Doesn’t Reduce Work—It Intensifies It</a><br/> </div><br/> <stream-item data-image="https://hbr.org/resources/images/article_assets/2026/02/Feb26_07_EynonJones-383x215.jpg" data-authors="Aruna Ranganathan" data-content-type="Digital Article" class="stream-item" data-topic="" data-id="tag:blogs.harvardbusiness.org_2007-03-31:999.422836" data-url="https://hbr.org/2026/02/ai-doesnt-reduce-work-it-intensifies-it" data-title="AI Doesn’t Reduce Work—It Intensifies It"><br/> </stream-item><br/> </div><br/> </li><br/> <li><br/> <div style="display: flex;"><br/> <figure class="product-thumb"><br/> <a rel="nofollow" href="https://hbr.org/2025/01/research-why-some-companies-weather-trade-wars-better-than-others?ab=at_art_art_rr_v1x4_s03"><br/> <img alt="" src="https://hbr.org/resources/images/article_assets/2025/01/Jan25_15_1787167835-383x215.jpg"><br/> </a><br/> </figure><br/> <div class="hed title-ellipses-3 title-h4" js-target="ellipse-product-title"><br/> <div class="at-topic"><br/> Risk Management<br/> </div><br/> <a rel="nofollow" href="https://hbr.org/2025/01/research-why-some-companies-weather-trade-wars-better-than-others?ab=at_art_art_rr_v1x4_s03">Research: Why Some Companies Weather Trade Wars Better Than Others</a><br/> </div><br/> <stream-item data-image="https://hbr.org/resources/images/article_assets/2025/01/Jan25_15_1787167835-383x215.jpg" data-authors="Di Fan" data-content-type="Digital Article" class="stream-item" data-topic="" data-id="tag:blogs.harvardbusiness.org_2007-03-31:999.400328" data-url="https://hbr.org/2025/01/research-why-some-companies-weather-trade-wars-better-than-others" data-title="Research: Why Some Companies Weather Trade Wars Better Than Others"><br/> </stream-item><br/> </div><br/> </li><br/> <li><br/> <div style="display: flex;"><br/> <figure class="product-thumb"><br/> <a rel="nofollow" href="https://hbr.org/podcast/2026/03/the-new-leadership-structures-that-unblock-innovation?ab=at_art_art_rr_v1x4_s04"><br/> <img alt="" src="https://hbr.org/resources/images/article_assets/2015/12/HBR-Ideacast-HP-feed.png"><br/> </a><br/> </figure><br/> <div class="hed title-ellipses-3 title-h4" js-target="ellipse-product-title"><br/> <div class="at-topic"><br/> Innovation<br/> </div><br/> <a rel="nofollow" href="https://hbr.org/podcast/2026/03/the-new-leadership-structures-that-unblock-innovation?ab=at_art_art_rr_v1x4_s04">The New Leadership Structures that Unblock Innovation</a><br/> </div><br/> <stream-item data-image="https://hbr.org/resources/images/article_assets/2015/12/HBR-Ideacast-HP-feed.png" data-authors="" data-content-type="Audio" class="stream-item" data-topic="" data-id="tag:blogs.harvardbusiness.org_2007-03-31:999.424662" data-url="https://hbr.org/podcast/2026/03/the-new-leadership-structures-that-unblock-innovation" data-title="The New Leadership Structures that Unblock Innovation"><br/> </stream-item><br/> </div><br/> </li></ul></div></div></div><div><div class="ecommerce-module"><div data-testid="container" data-analytics="placement-right_rail3"></div><h4 class="spacing-narrow">Readers Also Viewed These Items</h4><br/><ul><br/> <li><br/> <div><br/> <figure class="product-thumb"><br/> <a rel="nofollow" href="https://store.hbr.org/product/business-plan-development-harvard-managementor/5678C?ab=at_art_idp_v1x2_s01"><img alt="" src="https://cdn11.bigcommerce.com/s-yneuaokjib/images/stencil/250x260/products/25675/305824/5678C_500__99932.1715064604.png?c=2"></a><br/> </figure><br/> <div class="hed title-ellipses-3 title-h4" js-target="ellipse-product-title"><a rel="nofollow" href="https://store.hbr.org/product/business-plan-development-harvard-managementor/5678C?ab=at_art_idp_v1x2_s01">Business Plan Development: Harvard ManageMentor</a></div><br/> <div class="content-type">Tool</div><br/> <a rel="nofollow" class="standard-cta" href="https://store.hbr.org/product/business-plan-development-harvard-managementor/5678C?ab=at_art_idp_v1x2_s01" aria-label="Business Plan Development: Harvard ManageMentor buy now">Buy Now</a><br/> <stream-item data-image="https://cdn11.bigcommerce.com/s-yneuaokjib/images/stencil/250x260/products/25675/305824/5678C_500__99932.1715064604.png?c=2" data-authors="Harvard Business Publishing" data-content-type="Tool" class="stream-item" data-topic="" data-id="5678C-HTM-ENG" data-url="https://store.hbr.org/product/business-plan-development-harvard-managementor/5678C" data-title="Business Plan Development: Harvard ManageMentor"><br/> </stream-item><br/> </div><br/> </li><br/> <br/> <li><br/> <div><br/> <figure class="product-thumb"><br/> <a rel="nofollow" href="https://store.hbr.org/product/monopolies-and-tech-giants-tools-for-preparing-your-team-for-the-future/10445?ab=at_art_idp_v1x2_s02"><img alt="" src="https://cdn11.bigcommerce.com/s-yneuaokjib/images/stencil/250x260/products/74366/283047/10445E_500__06068.1713552400.png?c=2"></a><br/> </figure><br/> <div class="hed title-ellipses-3 title-h4" js-target="ellipse-product-title"><a rel="nofollow" href="https://store.hbr.org/product/monopolies-and-tech-giants-tools-for-preparing-your-team-for-the-future/10445?ab=at_art_idp_v1x2_s02">Monopolies and Tech Giants: Tools for Preparing Your Team for the Future</a></div><br/> <div class="content-type">Book</div><br/> <a rel="nofollow" class="standard-cta" href="https://store.hbr.org/product/monopolies-and-tech-giants-tools-for-preparing-your-team-for-the-future/10445?ab=at_art_idp_v1x2_s02" aria-label="Monopolies and Tech Giants: Tools for Preparing Your Team for the Future} buy now">Buy Now</a><br/> <stream-item data-image="https://cdn11.bigcommerce.com/s-yneuaokjib/images/stencil/250x260/products/74366/283047/10445E_500__06068.1713552400.png?c=2" data-authors="Harvard Business Review" data-content-type="Book" class="stream-item" data-topic="" data-id="10445E-KND-ENG" data-url="https://store.hbr.org/product/monopolies-and-tech-giants-tools-for-preparing-your-team-for-the-future/10445" data-title="Monopolies and Tech Giants: Tools for Preparing Your Team for the Future"><br/> </stream-item><br/> </div><br/> </li><br/></ul></div><div class="Standard_right-rail-related-topics__X_A2m"><div class="RelatedTopics_container__2Nkb4 RelatedTopics_standard__UaUNv"><span class="RelatedTopics_related-topics__zEjID RelatedTopics_standard__UaUNv">Read more on <a rel="nofollow" class="RelatedTopics_topic-link__u_bsi" href="/topic/subject/ai-and-machine-learning?ab=articlepage-topic"><strong>AI and machine learning</strong></a> or related topics <a rel="nofollow" class="RelatedTopics_topic-link__u_bsi" href="/topic/subject/organizational-decision-making?ab=articlepage-topic"><strong>Organizational decision making</strong></a>, <a rel="nofollow" class="RelatedTopics_topic-link__u_bsi" href="/topic/subject/generative-ai?ab=articlepage-topic"><strong>Generative AI</strong></a>, <a rel="nofollow" class="RelatedTopics_topic-link__u_bsi" href="/topic/subject/technology-and-analytics?ab=articlepage-topic"><strong>Technology and </strong></a></span></div></div></div></div></div>

22 more annotations...

• Having already fallen $800 billion short of turning a profit on the AI boom, some of the tech industry’s biggest players have accepted that the financial arithmetic on AI just doesn’t add up.
• These fellas argued that there are essentially two kinds of economic bubble: the good ones, like the Dot Com bubble, and the bad ones, like the 2008 subprime lending crisis.
• With bubbles, Hobart told the Atlantic, a “set of investments that you could never underwrite otherwise suddenly makes sense.”
• Yes, bubbles create volatility. Yes, investors lose money. Yes, employees lose jobs when companies fail. But the alternative is underinvestment in transformative opportunities.”
• OpenAI CEO Sam Altman has likewise argued that AI will a “huge net win for the economy” no matter what, even if a “phenomenal amount of money” ends up in the blender.
• . Some will go bankrupt, sure, but that just means less competition for market dominance. Each sees themselves as too big to fail — so when the bubble does burst, you can expect plenty of drama.
• “You should expect OpenAI to spend trillions of dollars on data center construction in the not very distant future,” Altman added. “You should expect a bunch of economists to wring their hands.”

5 more annotations...

• ​The text outlines a new landscape where HR’s value shifts from “owning processes” to becoming architects of human capability systems.
• Most of the HR models in use today were designed for industrial-era organisations built around human labour, hierarchical authority and slow information flows.
• We are entering the Fusion Era, where the workforce is no longer exclusively human, and intelligence is embedded directly into systems
• AI-centric organisations will not need HR departments as process managers but architects of human capability systems instead.
• In traditional organisations, information systems record activity after decisions are made, whereas in AI‑centric organisations, intelligence is embedded inside the workflow itself and systems do not simply store information, they interpret it, analyse it and when required, act on it.
• • Decision support becomes continuous rather than occasional.
      
  • Workflows become algorithmically orchestrated.
      
  • Employees interact directly with systems rather than through administrative intermediaries.
• In an AI‑centric organisation, the system executes the process, and the resulting implication is clear: if HR continues to define its value through owning processes, it is building its future on disappearing work.

• These capabilities are not there to support HR; they bypass it.

      
• • Human Judgement
   
      

  Strategic direction, ethical choices and organisational identity remain human responsibilities requiring context, accountability and moral reasoning. AI may well inform these choices, but it does not replace them.
• • AI‑Assisted Decisions
   
      

  In some operations, systems generate insights while humans validate the outcome. In this category we can include recruitment screening, workforce risk alerts, compensation benchmarks and learning recommendations. The system performs the analysis, the human makes the judgement.

      
   
• • Autonomous System Decisions
   
      

  Some decisions can become fully automated once reliability of the process has been established. Scheduling, workflow routing and resource allocation will increasingly operate without human intervention. The system does not advise, it acts.
• without governance they may hallucinate, misinterpret data or trigger unintended actions across enterprise workflows.

• Trust requires governance, and that governance can only come from:

      

  • clear access and control over who builds and deploys agents 
  • guardrails that restrict system behaviour 
  • decision logic audit mechanisms 
  • strong data governance 
  • the lifecycle management of AI systems 
  • usage cost monitoring  
  • continuous system oversight

• The principle is simple: Trust but verify.

      
• . The design question, therefore, instead of ‘How many employees do we need?’ becomes ‘What combination of human and machine capability produces the best outcome?’ In reality, hiring a human becomes a design decision.
• Developing and rolling out new technology in the middle of a product trajectory is a very expensive process, not only in execution but in marketing as well. It’s not difficult to envision some of the long-established ‘legacy’ players struggling to change course in a market that has become suddenly dynamic due to growth of single-module applications and the greater ease in linking those modules together when needed.

• future lies in having a role in the architecture of human capability systems viz:

      
   
  • human‑AI work design 
      
  • workforce capability architecture 
      
  • decision governance 
      
  • workforce ethics and trust
      
  • legislative compliance

15 more annotations...