Nexus Institute for Work and AI: Research Deep Dive

WRKdefined Podcast Network

Join us for in-depth conversations with leading researchers at the intersection of work and artificial intelligence. Our hosts explore cutting-edge research, practical applications, and the profound implications AI is having on the workplace. From automation and job transformation to ethical considerations and emerging opportunities, each episode unpacks complex topics through thoughtful dialogue with the experts shaping the future of work. Whether you're a researcher, practitioner, or simply curious about how AI is reshaping our professional lives, these deep dives offer valuable insights grounded in rigorous research and real-world experience.

  1. 8h ago

    Bridging the AI Dividend Gap

    The AI dividend gap describes the uneven distribution of generative artificial intelligence benefits in the workplace, where advantages flow primarily to college graduates, managers, and knowledge workers rather than frontline staff. Although research demonstrates that novices and lower-skilled employees often achieve the greatest productivity gains from artificial intelligence, regular tool usage remains concentrated among individuals who already hold high-quality jobs. To address this disparity, organizations must implement broad access, targeted training, and meaningful employee participation in technological decisions. Furthermore, leaders should establish formal agreements regarding gain-sharing and transparency while actively reinvesting productivity dividends into enhanced job roles. Ultimately, closing this divide transforms artificial intelligence from a source of workplace anxiety into an opportunity for shared organizational prosperity.

  2. 1d ago

    The AI Hiring Arms Race

    The modern recruitment landscape is experiencing an arms race as generative artificial intelligence floods employers with an overwhelming volume of polished applications while simultaneously fueling rising candidate fraud. This widespread technological adoption has diminished traditional application signals, eroded mutual trust between job seekers and hiring managers, and heightened legal scrutiny regarding automated screening bias. To navigate this challenge, organizations must abandon reactionary algorithmic filtering and instead implement evidence-based recruitment strategies. Key corrective measures include redesigning the application process around demonstrable skills, shifting evaluation weight toward high-validity assessments like structured interviews, and applying proportionate identity verification methods. Furthermore, enterprises need to establish strict internal governance for AI tools alongside transparent candidate policies that clearly outline acceptable AI usage. Ultimately, treating hiring as a rigorous measurement system allows businesses to cut through digital noise and restore long-term validity to talent selection.

  3. 3d ago

    The Cloned Voice at the Top

    This research investigates the growing threat of deepfake executive impersonation, wherein bad actors leverage generative artificial intelligence to convincingly clone the voices and appearances of corporate leaders. It emphasizes that the core vulnerability is hierarchical and cultural rather than technical, driven by an authority bias that makes employees hesitant to question requests originating from the top of the chain of command. Documented incidents across various global industries demonstrate that workers frequently prioritize rapid compliance over security due to fear of negative repercussions. To combat this rising wave of sophisticated financial fraud and unauthorized system access, organizations must cultivate a culture of verification supported by clear leadership norms. Ultimately, the literature advocates for a layered defense strategy featuring out-of-band verification protocols, strict payment controls, hardened identity workflows, and interactive practice to ensure that no single fraudulent voice can manipulate corporate assets.

  4. 3d ago

    Managing AI-Assisted Performance Reviews

    Generative artificial intelligence has quietly permeated workplace performance evaluations, as a vast majority of corporate managers utilize these tools while keeping employees largely unaware. This undisclosed ghostwriting creates a critical trust deficit, threatening organizational justice, exacerbating potential biases, and substituting machine fluency for genuine human insight. To safeguard interpersonal relationships and employee development, organizations must establish transparent disclosure norms and ensure that evaluative judgments remain strictly human and named. Furthermore, leaders should repurpose artificial intelligence strictly for preparation and rehearsal rather than text generation, while actively auditing for bias and adhering to emerging legal frameworks. Ultimately, organizations must harness these technologies to support continuous coaching rather than diminish the vital personal connection necessary for professional growth.

  5. 4d ago

    The Time Dividend Trap

    Generative artificial intelligence is frequently advertised as a tool that returns hours to employees, but empirical evidence reveals that these efficiency gains rarely translate into meaningful relaxation or slack. Instead, organizations and workers quickly absorb the saved minutes into expanded task portfolios, accelerated pacing, and heightened output expectations, creating a phenomenon termed the time dividend trap. Because companies typically capture these tech-driven time savings as default volume rather than deliberate allocation, employees often experience increased job demands, boundary erosion, and professional burnout. To counter this upward ratchet in workplace effort, leaders must intentionally measure net efficiency, establish clear reinvestment rules, and direct recovered hours toward human recovery and continuous learning. Ultimately, escaping this productivity trap requires recognizing that technology alone does not dictate how efficiency benefits are distributed; human choices and strategic governance shape the sustainable future of modern work.

  6. 6d ago

    Navigating the Shift from AI Users to Builders

    Employer demand for artificial intelligence capabilities is rapidly evolving from basic prompting toward hands-on tool construction and workflow automation. Although a vast majority of employees recognize the specific competencies they need to succeed in this changing landscape, a persistent support gap remains because many organizations fail to provide adequate developmental resources. To address this mismatch, companies must implement comprehensive strategies such as mapping role-specific capability rungs, funding dedicated learning time, and encouraging practical project builds. Furthermore, fostering collaborative networks through champions and engaged managers helps circulate vital knowledge across teams. Ultimately, leaders must align performance expectations with measurable outcomes and strategic judgment rather than superficial usage metrics to ensure sustainable workforce readiness.

  7. Oct 10

    Designing for Hybrid Intelligence

    Although organizations heavily invest in artificial intelligence, many fail to see meaningful returns because they mistakenly view technology as a human replacement rather than a complement. This research examine the concept of hybrid intelligence, which unites human and artificial intelligence to achieve outcomes superior to what either could accomplish independently. Successfully implementing this approach requires intentional task architecture, ensuring that humans and machines are assigned roles matching their respective strengths. Furthermore, systems must prioritize interpretability and transparency so that users can develop calibrated trust rather than falling into patterns of automation complacency. Ultimately, capturing the full economic and operational value of artificial intelligence depends on thoughtful workflow design and fostering a continuous co-learning process between people and machines.

  8. Oct 9

    Mental Models of AI Error Boundaries

    This research explores how human-AI collaboration relies heavily on the ability of people to build accurate mental models of artificial intelligence error boundaries rather than solely depending on algorithmic accuracy. When organizations deploy AI in critical fields like healthcare and criminal justice, team performance often falters because users struggle to recognize when a system will succeed or fail. To fix this, the literature argues that models should be designed with parsimonious and non-stochastic error patterns that make failure modes easy to understand and anticipate. Furthermore, organizations must implement structured training, clear uncertainty communication, and thoughtful software updates to prevent misplaced trust or unwarranted distrust in automated recommendations. Ultimately, successful deployment requires shifting the focus of development toward team performance and human-centered governance instead of prioritizing raw predictive metrics alone.

See All (185)

2
out of 5
2 Ratings

About

Join us for in-depth conversations with leading researchers at the intersection of work and artificial intelligence. Our hosts explore cutting-edge research, practical applications, and the profound implications AI is having on the workplace. From automation and job transformation to ethical considerations and emerging opportunities, each episode unpacks complex topics through thoughtful dialogue with the experts shaping the future of work. Whether you're a researcher, practitioner, or simply curious about how AI is reshaping our professional lives, these deep dives offer valuable insights grounded in rigorous research and real-world experience.

Information

You Might Also Like