CIO Agenda 2026: Delivering the AI Promise for Enterprise AI
CIO Agenda 2026: Delivering the AI Promise for Enterprise AI
Tim Crawford and Isaac Sacolick, both former Chief Information Officers and world-class CIO advisors, join Michael Krigsman on CXOTalk episode 909 to break down why enterprise AI strategies are failing, what separates transformational CIOs from those who are drowning, and why earning your seat at the table matters more than ever in 2026.
You'll discover:
✅ Why Tim says both AI strategy AND IT execution are failing, and what CIOs are focused on instead of outcomes
✅ The "three-legged race" framework: how CIO behavior, IT culture, and external perception must align for strategic credibility
✅ Why most CIOs have only a "layperson's understanding" of their own business, and how that kills AI value
✅ Tim's two swim lanes of AI success: invisible integration or robust training (there is no middle ground)
✅ Why Isaac says AI is "reshaping" business but not yet "transforming" it, and the product management shift that changes everything
✅ How to evaluate agentic AI: the human-in-the-loop vs. human-out-of-the-loop decision framework and why cybersecurity proves you can't wait
✅ The shadow AI paradox: why the best CIOs encourage it (with guardrails) instead of shutting it down
✅ The three skills every IT professional needs now: business acumen, critical thinking, and data literacy
⏱️ TIMESTAMPS
0:00 Cold open: "If you think you should have a seat at the table, you've failed"
0:35 Why both AI strategy and IT execution are failing
2:08 The productivity measurement problem with AI
2:45 What CEOs and boards want from CIOs in 2026
4:28 Why CIOs don't truly understand their business
6:54 Why organizations are stuck in AI pilot mode
9:04 Tim's 2 swim lanes: invisible AI vs. training-wrapped AI
11:23 Audience Q&A: Inside-out thinking vs. outside-in thinking
14:34 The 3-legged race: earning your seat at the table
17:09 Moving from AI efficiency to true business transformation
20:03 The shift from project-oriented to product-oriented IT
20:31 AI governance, CISO alignment, and data sensitivity
27:15 Agentic AI: fully autonomous vs. human-in-the-loop
34:46 Agentic AI strategy and the value equation (opportunity minus cost)
38:46 Shadow AI: innovation source or security threat?
43:00 Governance as culture, not a bolt-on
46:00 The AI skills gap: business acumen, critical thinking, data skills, and curiosity
49:46 Are survival-mode CIOs sabotaging their careers?
52:15 What CIO greatness looks like in 2026
🔔 SUBSCRIBE to CXOTalk for unfiltered conversations with the world's top technology and business leaders.
📩 Get notified about upcoming episodes. Subscribe to the CXOTalk newsletter: https://newsletter.cxotalk.com
🎙️ Read the summary and full transcript: https://www.cxotalk.com/episode/cio-agenda-2026-delivering-on-the-ai-promise
#CXOTalk #CIOAgenda2026 #AIStrategy #AgenticAI #DigitalTransformation #CIO #AIGovernance #EnterpriseAI #AILeadership #BusinessTransformation
CIO Agenda 2026: Delivering on the AI Promise
Learn why most organizations struggle to generate measurable value from AI and what CIOs must do differently to deliver real business results in CXOTalk episode 909.
Renowned MIT/Stanford computer scientist on AI and collective intelligence
MIT and Stanford professor Alex "Sandy" Pentland, one of the most cited researchers in the world with over 165,000 citations, explains why the real AI advantage isn't smarter models but collective intelligence. It's smarter humans working together with AI as the connective tissue. Drawing from his latest book Shared Wisdom, Pentland reveals the frameworks behind community intelligence and why data ownership, not frontier AI, will determine who wins the next decade.
You'll discover:
✅ Why "people plus AI" consistently beats AI alone, and the hedge fund evidence that proves it
✅ How "AI buddies" are replacing corporate manuals, newsletters, and hallway conversations to keep distributed teams aligned
✅ The Deliberation.io tool that makes meetings more than twice as effective by neutralizing power dynamics and keeping groups focused
✅ Why a 350,000-person multinational is cutting in-house staff to 150,000 while hiring 100,000 more project-based workers, and how AI enables that shift
✅ How a doctor with zero technical background built a hospital operating system in 6 weeks using AI tools
✅ The staggering stat: AI costs are dropping by 50% every 3.5 months, a factor of 1,000 over three years, and what that means for personal, on-device AI
✅ Why China's Belt and Road and India's Citizen Stack (1.4 billion customers signed up) are quietly winning the global data game while Silicon Valley focuses on frontier models
✅ Sandy's provocative proposal: a 10% equity contribution to sovereign wealth funds at company formation, which would have created a $10 trillion US fund if started in 1990
⏱️ TIMESTAMPS
0:00 Why AI alone loses money: the hedge fund reality check
2:07 Shared wisdom, community intelligence, and organizational culture
4:25 AI buddies: the brilliant librarian inside your company
5:44 Deliberation.io: making meetings 2x more effective
7:01 Using AI for exploration and long-range strategic thinking
11:29 Who's to blame when AI fails: executives or the machine?
14:28 Why AI can't do causality and what that means for leaders
18:14 AI's killer app for remote work and distributed organizations
21:09 A doctor built a hospital OS in 6 weeks: small teams, massive impact
24:09 Job displacement, social safety nets, and the sovereign wealth fund idea
27:01 Reinventing education: Costa Rica's bet and the MIT Media Lab model
32:16 LLMs vs. older AI: why you need both (and the loyalagents.org initiative)
37:13 Practical starting points for redesigning work with AI
40:16 Misinformation, data provenance, and the billion-dollar North Korea problem
48:50 The global data race: China, India, UAE, and why frontier models aren't the game
54:00 Cybersecurity warning: agentic AI creates massive new attack surfaces
🔔 Subscribe so you never miss a conversation with the world's top business and technology leaders.
📩 Get notified about upcoming shows. Sign up for the CXOTalk newsletter: https://newsletter.cxotalk.com
💬 Check the summary and full transcript: https://www.cxotalk.com/episode/ai-and-collective-intelligence-for-smarter-decision-making
🎙️ ABOUT CXOTALK
CXOTalk features unfiltered conversations with C-suite executives from major companies about AI, digital transformation, and business strategy. Hosted by Michael Krigsman.
Episode 907 | Recorded February 8, 2026
#CXOTalk #CollectiveIntelligence #AIStrategy #SandyPentland #CommunityIntelligence #SharedWisdom #EnterpriseAI #FutureOfWork #AILeadership #DigitalTransformation
Intelligent Orchestration: Better AI Coding and Software Delivery | GitLab CEO
AI coding tools are writing more code than ever, but your software isn't shipping any faster. Welcome to the AI Paradox and the solution, intelligent orchestration.
Bill Staples, CEO of GitLab, explains why AI-accelerated coding is actually creating massive downstream bottlenecks in code reviews, security checks, and deployment, and why adding more AI tools only makes the problem worse. GitLab's solution: intelligent orchestration across the entire software development lifecycle.
You'll discover:
✅ The "AI Paradox:" why faster coding isn't translating into faster software delivery
✅ How tool fragmentation and context-switching are killing developer productivity
✅ Why agents that thrive on context fail when your tools are siloed
✅ The "inner loop architecture" that makes AI agents 40% more accurate and 25% faster
✅ How GitLab's intelligent orchestration approach combines workflows, context, and guardrails
✅ Why mid-level developers are about to become strategic orchestrators (not just coders)
✅ The exact metrics CIOs should track, and why "lines of code" is the wrong one
✅ First steps: audit, consolidate, and pilot before going all-in on AI
⏱️ TIMESTAMPS
0:00 The AI Paradox: Why faster coding doesn't mean faster delivery
1:10 How tool fragmentation creates developer bottlenecks
3:40 Why AI agents make complexity worse (not better)
5:12 Solving the AI automation problem: people, process, and technology
6:36 Inner loop architecture: co-locating agents and data
9:14 Intelligent orchestration: workflows, context, and guardrails
10:32 How GitLab's knowledge graph supercharges agent accuracy
12:49 Universal guardrails for humans and AI agents
13:39 Real-world results: 2-3x more merge requests, pipeline fixes in minutes
15:00 Common threads driving customer success
16:36 How AI transforms the mid-level developer's role
19:06 Advice for CIOs and CTOs putting this into practice
20:49 First steps: audit, measure, and pilot
22:45 Core metrics to evaluate AI's real value
25:02 Wrap-up
🔔 Subscribe for weekly conversations with top technology executives.
📩 Get the CXOTalk newsletter: https://newsletter.cxotalk.com
💬 Drop your questions in the comments: Michael and our community actively engage.
🎙️Read the summary: https://www.cxotalk.com/episode/intelligent-orchestration-software-delivery-for-the-ai-era-with-ceo-of-gitlab
#AI #SoftwareDevelopment #DevOps #GitLab #AIAgents #DeveloperProductivity #CIO #CTO #DigitalTransformation #CXOTalk
AI and Collective Intelligence for Smarter Decision-Making
In CXOTalk episode 907, renowned MIT/Stanford computer scientist Sandy Pentland explains how AI helps leaders make better decisions by improving information flow, building trust, and enabling more intentional collaboration.
Enterprise AI at Scale: How U.S. Bank's Chief AI Officer Deploys AI Across 70,000 Employees
Prashant Mehrotra, Chief AI Officer at US Bank, discusses how the bank evaluates AI initiatives and scales projects from pilot to production. He explains how to build customer trust through responsible AI design and prepare for the future of autonomous banking in CXOTalk episode 906. This conversation covers key aspects of AI in business and AI implementation within a large banking institution.
=======
Please support our sponsor Emeritus:
Explore executive education programs from Emeritus, in collaboration with top universities: https://cxotalk.partner.emeritus.org/
=======
Key topics discussed:
→ Why AI should transform processes, not simply make them more efficient
→ How U.S. Bank cut governance approval times in half by engaging risk partners early
→ The critical role of baselines in determining whether AI pilots scale or fail
→ Why "AI without data is a hallucination" and how the bank organizes Digital, Data, and AI under one leader
→ Building AI literacy across the entire workforce, from executives to frontline associates
→ The shift from building models to leveraging external foundation models at scale
→ Balancing personalization with privacy in customer interactions
Mehrotra emphasizes that the client remains the "North Star" for every AI initiative. He offers practical guidance on metrics, funding pilots through to production, and creating repeatable governance processes that accelerate rather than slow down AI deployment.
🔷 Show notes: https://www.cxotalk.com/episode/u-s-banks-chief-ai-officer-on-strategy-governance-and-scaling-ai
🔷 Newsletter: www.cxotalk.com/subscribe
🔷 LinkedIn: www.linkedin.com/company/cxotalk
🔷 Twitter: twitter.com/cxotalk
📍 CHAPTERS:
00:00 Introduction: AI at U.S. Bank
00:30 How AI Ideas Get Evaluated and Funded
02:24 The Chief AI Officer Role: The 4 E's Framework
05:21 AI as a Transformative Force, Not Just Technology
09:05 Real-World AI Impact: Customer Service and Developer Tools
11:15 Measuring AI Success: Metrics That Matter
15:31 Why 95% of AI Pilots Fail: The Importance of Baselines
17:05 AI and Sustainability Goals
18:01 Managing Risk and Governance at Scale
22:46 Moving from Pilots to Production
25:54 Aligning AI Strategy with Business Objectives
28:39 Building an AI-Ready Workforce: Skills and Education
32:12 How AI Differs from ERP and Traditional Technology Rollouts
35:05 Regulatory Compliance in AI-Driven Customer Outreach
37:16 Leadership Structure and Cross-Functional Collaboration
39:07 Data as the Foundation of AI
42:20 Technology Platforms and Vendor Partnerships
43:24 Build vs. Buy: Leveraging Foundation Models
45:56 AI as Strategic Investment: Creating Durable Value
47:58 Personalization Without Being Invasive
50:14 Conclusion
#AI #ArtificialIntelligence #Banking #DigitalTransformation #EnterpriseAI #ChiefAIOfficer #Leadership #CXOTalk
Robots: Why Physical AI Can't Make Errors (with CTO of Hexagon Robotics)
An AI chatbot that hallucinates is annoying. A robot or physical AI that hallucinates can cause injury or death. Burkhard Boeckem, CTO of Hexagon, explains why the bar for physical AI is fundamentally higher than digital AI, and what it takes to deploy robots that actually work in the real world. All this in CXOTalk episode 905.
In this conversation, we cover:
→ What physical AI actually means (and why it's different from the AI you use every day)
→ Why digital twins are the foundation for training robots safely
→ The gap between impressive YouTube demos and robots that create economic value
→ Functional safety: the "big theme" coming in 2026
→ Cloud vs. edge computing for autonomous systems
→ Where robotics deployments fail (hint: it's not the technology)
→ What boards get wrong about robotics investments
→ Timeline: when will we see real autonomy?
Key insight: "Many boards overestimate the speed and underestimate the system work. It's not a software rollout—it's a complex engineering system."
Burkhard's prediction: Autonomy in constrained environments is 1-3 years away. The "butler humanoid" that does everything? Still a ways off.
⏱️ CHAPTERS
00:00 Introduction: What is Physical AI?
02:06 Digital Twins as the Foundation
03:23 Understanding Ground Truth
06:42 Digital AI vs. Physical AI: Safety and Reliability
08:47 Real-World Business Applications
10:48 Security and Functional Safety
14:57 CES Announcements and Industry State
20:01 Cloud vs. Edge Computing in Robotics
22:32 Regulations for Physical AI
25:10 Addressing Bias in Physical AI
27:51 Timeline to Autonomy
31:46 Creating Economic Value Beyond Demos
33:26 Where Robotics Deployments Fail
35:59 The Future of Humanoid Form Factors
38:38 The Humanoid as User Interface
40:06 Digital Twins for Robotics
42:36 Fleet Collaboration and Swarm Intelligence
43:46 What Boards Get Wrong About Robotics
45:18 The Future of Work
46:32 Responsible Deployment
47:15 Manager AIs for Worker AIs?
48:16 Looking Ahead: Next 2-3 Years
49:37 Core Technical Challenges
————————————————
🔷 Full episode, show notes and resources: https://www.cxotalk.com/episode/google-clouds-cto-inside-the-ai-strategy
🔷 Newsletter: www.cxotalk.com/subscribe
🔷 LinkedIn: www.linkedin.com/company/cxotalk
🔷 Twitter: twitter.com/cxotalk
————————————————
About the Guest:
Burkhard Boeckem is Chief Technology Officer at Hexagon, a global leader in digital reality solutions combining sensor, software, and autonomous technologies.
About CXOTalk:
CXOTalk features direct, unfiltered conversations with the world's top business and technology leaders. No fluff, no PR speak—just honest discussion about what's actually working (and what isn't).
#PhysicalAI #Robotics #Humanoids #DigitalTwin #AI #AutonomousSystems #CXOTalk #EnterpriseAI #Manufacturing #FutureOfWork
US Bank's Chief AI Officer on Strategy, Governance, and Scaling AI
Prashant Mehrotra, Chief AI Officer at US Bank, explains how the bank evaluates AI initiatives, scales projects from pilot to production, builds customer trust through responsible AI design, and prepares for the future of autonomous banking, on CXOTalk episode 906.
Robots: Why Physical AI Can't Be Wrong | CTO of Hexagon Robotics
An AI chatbot that hallucinates is annoying. A robot or physical AI that hallucinates can cause injury or death. Burkhard Boeckem, CTO of Hexagon, explains why the bar for physical AI is fundamentally higher than digital AI, and what it takes to deploy robots that actually work in the real world. All this in CXOTalk episode 905.
In this conversation, we cover:
→ What physical AI actually means (and why it's different from the AI you use every day)
→ Why digital twins are the foundation for training robots safely
→ The gap between impressive YouTube demos and robots that create economic value
→ Functional safety: the "big theme" coming in 2026
→ Cloud vs. edge computing for autonomous systems
→ Where robotics deployments fail (hint: it's not the technology)
→ What boards get wrong about robotics investments
→ Timeline: when will we see real autonomy?
Key insight: "Many boards overestimate the speed and underestimate the system work. It's not a software rollout—it's a complex engineering system."
Burkhard's prediction: Autonomy in constrained environments is 1-3 years away. The "butler humanoid" that does everything? Still a ways off.
⏱️ CHAPTERS
00:00 Introduction: What is Physical AI?
02:06 Digital Twins as the Foundation
03:23 Understanding Ground Truth
06:42 Digital AI vs. Physical AI: Safety and Reliability
08:47 Real-World Business Applications
10:48 Security and Functional Safety
14:57 CES Announcements and Industry State
20:01 Cloud vs. Edge Computing in Robotics
22:32 Regulations for Physical AI
25:10 Addressing Bias in Physical AI
27:51 Timeline to Autonomy
31:46 Creating Economic Value Beyond Demos
33:26 Where Robotics Deployments Fail
35:59 The Future of Humanoid Form Factors
38:38 The Humanoid as User Interface
40:06 Digital Twins for Robotics
42:36 Fleet Collaboration and Swarm Intelligence
43:46 What Boards Get Wrong About Robotics
45:18 The Future of Work
46:32 Responsible Deployment
47:15 Manager AIs for Worker AIs?
48:16 Looking Ahead: Next 2-3 Years
49:37 Core Technical Challenges
————————————————
🔷 Full episode, show notes and resources: https://www.cxotalk.com/episode/google-clouds-cto-inside-the-ai-strategy
🔷 Newsletter: www.cxotalk.com/subscribe
🔷 LinkedIn: www.linkedin.com/company/cxotalk
🔷 Twitter: twitter.com/cxotalk
————————————————
About the Guest:
Burkhard Boeckem is Chief Technology Officer at Hexagon, a global leader in digital reality solutions combining sensor, software, and autonomous technologies.
About CXOTalk:
CXOTalk features direct, unfiltered conversations with the world's top business and technology leaders. No fluff, no PR speak—just honest discussion about what's actually working (and what isn't).
#PhysicalAI #Robotics #Humanoids #DigitalTwin #AI #AutonomousSystems #CXOTalk #EnterpriseAI #Manufacturing #FutureOfWork
Robot Realities & Physical AI
Hexagon CTO Burkhard Boeckem separates hype from reality on physical AI, robotics, and digital twins, exploring what it actually takes to deploy autonomous systems when failure isn't an option, on CXOTalk episode 904.
Cybersecurity and Quantum Computing: A Readiness Guide (with Palo Alto Networks)
Nation-states are already harvesting your encrypted data, betting they'll crack it when quantum computers mature. It's happening now. In CXOTalk episode 904, Anand Oswal, Executive Vice President of Network Security at Palo Alto Networks, explains what business leaders need to understand about the quantum security and how to address it. The impact of quantum computing on cybersecurity will be huge.
Key timelines discussed:
-- Cryptographically relevant quantum computers expected by end of this decade
-- RSA and ECC algorithms deprecated by 2030, disallowed by 2035
-- Enterprise cryptographic migrations typically take 5-10 years
🔷 Show notes: https://www.cxotalk.com/episode/cybersecurity-and-quantum-computing-a-readiness-guide
🔷 Newsletter: www.cxotalk.com/subscribe
🔷 LinkedIn: www.linkedin.com/company/cxotalk
🔷 Twitter: twitter.com/cxotalk
Chapters:
00:00 Introduction
00:36 What is quantum computing?
01:17 The core security threat
02:29 Customer awareness of quantum risk
02:57 Timeline: When will quantum arrive?
03:53 Implications for security leaders
05:35 Compliance timelines
06:24 Business risk explained
07:52 How encryption works today
09:22 Harvest now, decrypt later attacks
09:57 How the industry is preparing
12:27 Pillar 1: Discovery and visibility
14:26 Pillar 2: Protect
16:29 Pillar 3: Accelerate
17:40 The role of AI
18:10 Talent and skills
19:31 Practical steps for quantum readiness
20:43 What leaders should do now
Real World AI Agents: Inside Google's Strategy, with WIll Grannis, CTO, Google Cloud
Google is betting billions on AI agents and multimodal models. Will Grannis, CTO of Google Cloud, works with their top 150 enterprise customers on AI implementation. Here's what he's seeing.
In this conversation:
Why agents are the "third wave of automation"
Real examples: A UK telco using multimodal AI for customer support
How to architect AI evaluation at scale
The future of multimodal AI (Nano Banana, Veo, Gemini Live)
How to separate AI hype from reality
What's coming in the next 6-18 months
Why CEOs can, and should, build agents themselves
🔷 Full episode, show notes and resources: https://www.cxotalk.com/episode/google-clouds-cto-inside-the-ai-strategy
🔷 Newsletter: www.cxotalk.com/subscribe
🔷 LinkedIn: www.linkedin.com/company/cxotalk
🔷 Twitter: twitter.com/cxotalk
Error 400: API key not valid. Please pass a valid API key..
Domain code: global
Reason code: badRequest
