In this HCI Webinar, Dr. Jonathan H. Westover talks with Alexandria Agresta about why the workplace needs more fun! Disco balls. DJ decks. Dancing queens. Alexandria Agresta brings it all. As the World’s First DJing Speaker, she fuses the insight of a TED Talk with the electricity of a music festival to deliver her groundbreaking keynote, The Business Party. Her mission is simple: to spark the next wave of bold leadership and transform the workplace into a WOWplace, where possibility, creativity, and fun take center stage. Now, let’s get this party started!

In this podcast episode, Dr. Jonathan H. Westover talks with Alexandria Agresta about why the workplace needs more fun! Disco balls. DJ decks. Dancing queens. Alexandria Agresta brings it all. As the World’s First DJing Speaker, she fuses the insight of a TED Talk with the electricity of a music festival to deliver her groundbreaking keynote, The Business Party. Her mission is simple: to spark the next wave of bold leadership and transform the workplace into a WOWplace, where possibility, creativity, and fun take center stage. Now, let’s get this party started! Check out all of the podcasts in the HCI Podcast Network (https://www.podbean.com/podcast-network/HCI) !

Artificial intelligence (AI) is transforming the workplace at an unprecedented speed, yet a significant challenge known as the AI capability gap has emerged. This gap refers to the disconnect between the advanced, generalized AI tools developed by engineers and data scientists, and the specialized, domain-specific needs of professionals in fields such as healthcare, finance, and law. While AI tools possess remarkable speed and power, they lack the nuanced understanding required to apply these tools effectively in complex, regulated environments. Highlights 🤖 The AI capability gap is the disconnect between powerful AI tools and the specialized knowledge needed to use them effectively. 🚀 Organizations empowering domain experts to experiment with AI see faster innovation and long-term competitive advantages. ⏳ Allocating as little as 10% of work time for AI experimentation fosters discovery of high-value applications. 📚 Peer-led, context-specific training is more effective than generic tutorials for AI adoption. 🌉 AI translators—domain experts with AI knowledge—are critical in bridging technical and business teams. ⚠️ Ignoring the AI capability gap risks productivity loss, compliance errors, and career stagnation. 🏆 Leadership commitment and a culture of experimentation are essential to closing the AI capability gap and driving innovation. Key Insights 🤖 The AI capability gap is a fundamental challenge in AI adoption: AI tools are designed to be general-purpose and powerful but lack the domain-specific contextual understanding necessary for effective application. This gap highlights the importance of integrating domain expertise with AI capabilities rather than relying solely on technical development. Companies must recognize that AI is not plug-and-play; its value lies in tailored application driven by those who understand the nuances of their work. 🚀 Empowering domain experts to experiment accelerates innovation: Case studies consistently show that organizations allowing professionals to dedicate time to AI experimentation uncover better use cases and innovate faster. This approach contrasts with passive adoption, where organizations lag behind competitors and struggle to catch up. Encouraging safe, deliberate experimentation enables continuous learning and the development of best practices that compound over time. ⏳ Time allocation for AI learning signals organizational priorities: Dedicating even a modest portion of the workweek (e.g., 10%, or about four hours) for AI exploration creates psychological permission for learning and innovation. This structured time encourages employees to move beyond urgent deadlines and discover AI applications that yield significant productivity and quality improvements. Without this dedicated time, AI adoption remains superficial and fragmented. 📚 Context-specific, peer-led training enhances adoption and relevance: Generic AI tutorials often fail to resonate with professionals because they lack relevance to daily challenges. Training led by peers who share real-world successes and failures fosters a shared language and culture around AI use. This practical approach helps build confidence, encourages adoption, and accelerates the diffusion of effective AI workflows within teams and organizations. 🌉 AI translators are key to bridging the gap between technology and business: These hybrid roles combine domain expertise with a deep understanding of AI capabilities and limitations. AI translators identify promising projects, define requirements, and ensure that AI solutions address real business problems. They act as crucial intermediaries, preventing misaligned or inefficient AI deployments and maximizing return on investment. ⚠️ Ignoring the AI capability gap has severe consequences: Organizations that neglect AI fluency risk productivity stagnation, missed insights, compliance breaches, reputational damage, and costly error remediation. Similarly, individuals who avoid developing AI skills face shrinking roles and potential job displacement as routine tasks become automated. The stakes are high, making proactive AI learning not just beneficial but essential for future relevance. 🏆 Leadership support and culture shape successful AI integration: Executive commitment is vital for closing the AI capability gap. This includes allocating budgets for training and experimentation, rewarding AI skills development, and fostering a culture that values curiosity and psychological safety. Leaders who champion AI fluency create an environment where failure is a learning opportunity, enabling the workforce to innovate and adapt rapidly. This cultural foundation transforms AI from a tool into a strategic engine for growth. Perfect for leaders, practitioners, and innovators looking to turn AI from a tool into a strategic capability. Like/share if this helped—spread the playbook. #AICapabilityGap #AIFluency #AIIntegration #DomainExperts #FirstMoverAdvantage #OrganizationalLearning

Abstract: Artificial intelligence presents organizations with an unprecedented paradox: the engineers building AI systems possess limited insight into optimal applications within specific professional domains, while domain experts often lack the technical fluency to unlock AI's potential in their fields. This capability gap creates a strategic window for practitioners who bridge both worlds—combining deep domain knowledge with AI literacy—to establish competitive advantages before commoditization occurs. This article examines the structural reasons behind this expertise divergence, quantifies the organizational stakes of the capability race, and provides evidence-based frameworks for domain experts to systematically discover, validate, and institutionalize high-value AI applications. Drawing on innovation diffusion research, organizational learning theory, and documented cases across healthcare, legal services, and financial analysis, we demonstrate that first-mover advantages in AI application development yield compounding returns through proprietary workflow optimization, talent retention, and market repositioning. The analysis concludes with actionable strategies for building durable AI capabilities that transcend tool adoption to fundamentally reshape competitive dynamics within professional fields.

Abstract: Despite $30–40 billion in enterprise GenAI investment, 95% of organizations achieve zero measurable return, trapped on the wrong side of what we term the "GenAI Divide." This review synthesizes findings from MIT's Project NANDA research examining 300+ AI implementations and interviews with 52 organizations to identify why pilots stall and how exceptional performers succeed. The divide stems not from model quality or regulation, but from a fundamental learning gap: most enterprise AI systems lack memory, contextual adaptation, and continuous improvement capabilities. While consumer tools like ChatGPT achieve 80% exploration rates, custom enterprise solutions suffer 95% pilot-to-production failure rates. Organizations crossing the divide share three patterns: they partner rather than build (achieving 2x higher success rates), empower distributed adoption over centralized control, and demand learning-capable systems that integrate deeply into workflows. Back-office automation delivers superior ROI compared to heavily-funded sales functions, though measurement challenges persist. The emerging agentic web—enabled by protocols supporting persistent memory and autonomous coordination—represents the infrastructure required to bridge this divide at scale.

Companies worldwide are investing billions in generative AI, anticipating significant boosts in productivity and innovation. However, the vast majority of these investments fail to yield substantial returns, with about 95% of enterprise AI pilots never advancing to full production. This stark disparity creates a divide between organizations experimenting with AI and those achieving tangible business impact. The core reason behind this divide is not merely technical but rooted in a fundamental learning gap within enterprise AI systems. These systems often lack persistent memory, fail to improve from user feedback, and cannot adapt effectively to dynamic workflows, which severely limits their ability to scale beyond pilot phases. Highlights 🤖 95% of enterprise generative AI pilots fail to reach full production, revealing a major adoption challenge. 🧠 The core issue is a fundamental learning gap: AI systems often lack memory, feedback loops, and workflow adaptation. 📉 Only 5% of AI projects starting at exploration make it to full-scale deployment, highlighting the pilot-to-production chasm. 🔄 Back-office automation provides the most consistent, measurable ROI compared to high-profile customer-facing AI applications. 👥 Over 90% of knowledge workers use consumer AI tools unofficially, indicating gaps in enterprise AI offerings. 🤝 Partnering with specialized AI vendors nearly doubles the success rate compared to building AI solutions internally. 🚀 Distributed AI experimentation across teams uncovers practical use cases faster than centralized labs. Key Insights 🧩 Learning Gap as the Central Barrier: The failure of AI pilots to scale is predominantly due to enterprise AI systems’ inability to learn continuously. Without persistent memory, AI tools forget past interactions, making them ineffective in building user context or improving over time. This gap prevents AI from evolving into smarter assistants, limiting their practical utility and user adoption in complex workflows. Organizations must prioritize AI solutions with learning capabilities to overcome this barrier. 🔄 Broken Feedback Loops Limit Improvement: Effective AI systems require mechanisms for users to provide feedback and corrections that the system can incorporate. Many enterprise pilots fail because there is no easy way for users to correct errors or guide the AI outputs. This rigidity causes repeated mistakes and user frustration, pushing employees toward more flexible consumer AI tools like ChatGPT. Integrating seamless, user-driven feedback into AI workflows is essential to improve accuracy and trust. 🛠️ Workflow Adaptation is Vital: AI must fit naturally into the dynamic and fluid nature of real-world work. Successful AI tools learn team-specific patterns, exceptions, and historical resolutions to become genuinely helpful. For example, an AI tool for IT support should adapt recommendations based on past ticket outcomes and user groups. Without this adaptability, AI solutions remain rigid and disconnected from actual operational needs, undermining adoption and value generation. 💼 Back-Office Automation Delivers Tangible ROI: While customer-facing AI applications attract attention, the most reliable and measurable benefits come from automating back-office tasks such as invoice processing, compliance checks, and internal support. These processes are critical to business operations and well-suited to AI-driven efficiency gains. Prioritizing these areas allows organizations to realize early wins, build momentum, and develop internal AI capabilities before tackling more complex initiatives. 👥 Shadow AI Reveals User Needs and Pain Points: The widespread unofficial use of consumer AI tools by employees—referred to as shadow AI—signals a clear demand for AI solutions that enterprise tools fail to meet. Rather than viewing shadow AI as a compliance problem alone, leaders should analyze its usage patterns to identify unmet needs, pain points, and potential pilot projects that align with actual user behavior and preferences. 🤝 Partnering Accelerates AI Success: Companies partnering with specialized AI vendors enjoy nearly double the success rate in crossing the generative AI divide compared to those building solutions internally. Vendors bring domain expertise, industry-specific training data, and proven deployment experience, enabling faster delivery of refined AI products. This approach allows organizations to leverage AI effectively without diverting excessive resources toward building foundational models and infrastructure. #GenAI #EnterpriseAI #AIAdoption #AITransformation #AgenticWeb OUTLINE: 00:00:00 - The Great GenAI Divide 00:01:08 - Why Most AI Projects Stall 00:02:55 - Memory, Feedback, and Adaptation 00:04:34 - The Back Office, Shadow AI, and the Agentic Future 00:06:09 - Cross the Divide Before It’s Too Late 00:07:31 - Partner, Distribute, and Demand Learning 00:09:10 - Deploy What Learns, Integrates, and Scales

In this HCI Webinar, Dr. Jonathan H. Westover talks with Jennifer Carlson about the role of apprenticeships in preparing the future workforce. Jennifer Carlson serves as the CEO of Apprenti. Apprenti is a non-profit, apprenticeship intermediary and workforce consulting organization that delivers a secondary pipeline of tech talent to address U.S. domestic digital skills shortages. A former business leader with AIG, Progressive and adjunct professor at Seattle University, Jennifer also serves on the Tech Councils of North America (TECNA) foundation board, and as an Advisory Board Member - Apprenticeships for America.

The video, authored by Dr. Jonathan H. Westover, explores the rapidly evolving role of artificial intelligence (AI) in education and how it can be integrated thoughtfully to enhance rather than diminish human learning capabilities. It opens by acknowledging the widespread and growing use of AI tools by students worldwide, highlighting the challenge this presents to educators: how to harness AI without undermining the fundamental goal of education, which is to develop independent thinking and critical skills. Highlights 🤖 AI tools are increasingly used by over half of high school and university students worldwide. ⚠️ AI-generated work can create a dangerous illusion of competence by bypassing critical learning processes. 🤝 A balanced approach treats AI as a learning partner, supporting but not replacing student thinking. 🧠 AI can help reduce cognitive load, enabling students to focus on higher-order thinking and metacognition. 📚 Assessments must shift from final products to portfolios that capture the learning journey, including drafts and reflections. 🌍 Addressing the digital divide and teaching AI literacy are crucial for equitable AI integration. 🔒 Data privacy concerns arise from student interactions with AI tools owned by major tech companies. Key Insights 🤖 Widespread AI adoption is transforming education rapidly: The fact that over half of high school and university students already use AI tools daily shows that AI is no longer a niche or future issue but a present reality in education. This transformation demands immediate and thoughtful responses from educators, policymakers, and institutions. Ignoring or resisting AI adoption is futile; instead, systems must adapt to this new norm. ⚠️ Illusion of competence threatens authentic learning: When AI completes tasks like writing essays or generating code, students may receive high grades without engaging in essential cognitive processes such as research, critical analysis, drafting, and revision. This risks producing graduates who can produce polished outputs but lack deep understanding and independent problem-solving skills. 🤝 AI as a partner rather than a replacement encourages deeper learning: Rather than banning AI or allowing it to do all the work, positioning AI as a supportive assistant empowers students to focus on complex thinking tasks. For example, automating citation formatting or summarizing information frees mental resources for evaluating arguments, synthesizing ideas, and reflecting on learning. 🧠 Metacognition is a critical skill enhanced by AI: Using AI tools to teach students how to think about their own thinking—metacognition—can profoundly improve learning outcomes. AI can scaffold this process by prompting students to reflect on their reasoning, question their assumptions, and revise their work thoughtfully. 📚 Redesigning assessment to focus on learning processes is essential: Traditional grading based solely on final products is inadequate in an AI-augmented environment. Portfolio-based assessments that include brainstorming sessions, multiple drafts with documented revisions, AI prompt histories, reflective writings, and peer feedback provide a richer, more transparent picture of student learning. 🌍 Equity and access must be at the forefront of AI integration: The digital divide poses a significant barrier; students without reliable internet, modern devices, or quiet study spaces are disadvantaged. Furthermore, disparities in AI literacy mean some students can leverage AI tools more effectively than others. 🔒 Ethical considerations, especially data privacy, are paramount: Many AI tools collect and store sensitive student data on corporate servers, raising concerns about privacy, consent, and data security. Educational institutions must advocate for transparent data policies, protect student information, and educate students about the implications of sharing personal data with AI platforms. If this helped, please like and share the video. #AIinEducation #EdTech #Metacognition #AssessmentReform #JonathanWestover #TeachingWithAI OUTLINE: 00:00:00 - Intro — Title & Table of Contents 00:00:39 - The AI Challenge in Our Classrooms 00:01:34 - The Risk of Replacing Thought 00:02:39 - AI as a Learning Partner 00:03:35 - Redesigning Our Classrooms and Assessments 00:04:38 - Building an Ethical and Equitable AI Future