This research note examines how artificial intelligence is changing the relationship between early-career work, learning and capability development. Drawing on five structured conversations with early-career professionals and one strategic peer conversation, it identifies three emerging organizational signals: developmental tasks may be automated while low-learning work remains; stronger AI-assisted outputs may conceal weaker understanding and ownership; and broad responsibility does not necessarily translate into meaningful development.

The note argues that sustainable AI-enabled productivity requires organizations to redesign work, learning and career progression together. It proposes practical questions for deciding which tasks should be automated, augmented, protected or redesigned, and how employees should be expected to verify, explain, judge and independently modify AI-assisted work.

The findings are exploratory and are intended to establish a research agenda rather than make representative claims about Gen Z as a whole.

Emerging Signals on Gen Z, AI and Early-Career Development

The Missing First Rung

What readers will get
A comparative view of how AI in education is unfolding across the United States, Kenya, China, the United Arab Emirates and Switzerland.
A practical map of the main implementation bottlenecks: teacher readiness, policy clarity, trust, infrastructure and assessment reform.
Concrete signals for AI practitioners, education leaders and policy actors seeking grounded lessons rather than hype.

Legacy education-era release preserved in the archive.

Signals, bottlenecks, and practical moves for AI practitioners, educators, and policy leaders

AI in Education Across Five Countries: Implementation Signals, Policy Gaps, and What Comes Next

Artificial intelligence is transforming education, but its benefits remain out of reach for many of the communities that could benefit most. Drawing primarily from an hour-long public interview with Dr. Seiji Isotani and secondarily from the OECD Digital Education Outlook 2026 interview chapter, this memo argues that policymakers should stop treating infrastructure build-out as a precondition for AI in education. Instead, they should design around the infrastructure that already exists—especially mobile phones, intermittent connectivity, and teacher-led delivery models. The Brazil case discussed by Dr. Isotani shows that this approach can work at scale: 500,000 students across 7,000 schools and 20,000 teachers received materially faster feedback on writing, with statistically significant improvement and no meaningful urban-rural or resource-based gap in gains.

Based primarily on an policy dialogue with Dr. Seiji Isotani

AI in Education Unplugged: Closing the Access Gap Through Education-Driven Design

This brief synthesizes evidence and expert dialogue on scaling AI literacy and governance beyond pilots. It argues for treating AI literacy as baseline infrastructure, aligning responsible-use policy with cybersecurity maturity, and preparing for workforce change as task-displacement first.

Insights from a dialogue with Pat Yongpradit (Microsoft) + selective OECD/WEF evidence

Exiting the AI Pilot Trap: Responsible Scale for AI Literacy, Governance, and Workforce Readiness

AIGW Workforce Transition Case Study 01 - Salesforce and Agentforce

How Salesforce Is Reshaping Its Workforce in the Age of AI

Marc Benioff interview - The Logan Bartlett Show

How Salesforce Is Reimagining Its Workforce for the Agentic Enterprise

FY26 Stakeholder Impact Report

Building the Agentic Enterprise: AI Playbook Part 1

One transition, several workforce mechanisms

A human-agent operating model

Customer support became a human-agent operating model

Salesforce says it deployed Agentforce on help.salesforce.com to answer customer-support questions.

Agentforce enters customer support

Benioff described comparable customer satisfaction and support headcount moving from about 9,000 to about 5,000 because fewer support staff were needed.

Benioff reports lower staffing need

Salesforce reported 2.6 million Agentforce conversations, 63% resolution, hundreds redeployed, and a support team change involving attrition and redeployment.

Workforce mechanism becomes clearer

Salesforce described individual applications, cohort moves, targeted matching, work-time learning and non-backfilling as part of the transition.

Mobility and training pathways documented

From deployment to workforce redesign

After more than eight years in technical support, Chuck used Career Connect and additional training before moving into a role that bridges customers and product development.

Kristen Chuck -> Success Architect

Piludaria moved from Marketing Cloud support into Renewals and received extended training in a new commercial vocabulary and practice environment.

Ankita Piludaria -> Renewals

Thakkar moved from developer support engineering into Success Architecture, illustrating how technical and customer expertise could be reused rather than discarded.

Sarah Thakkar -> Success Architect

Domain expertise became transition capital

Redeploy | Reskill | Reduce

Strong workflow outcomes, bounded workforce attribution

Build mobility before task compression becomes a crisis.

Measure the handoff, not only the automation rate.

Employment effects may emerge through mechanisms standard layoff statistics miss.

What this case changes in the decision

How AI changed customer support, redeployed expertise, and reduced staffing requirements

Salesforce provides an unusually complete public view of how an AI deployment can change both work and workforce structure. Agentforce absorbed a large share of routine support interactions while human work shifted toward complex, proactive and consultative activity.

案例库

Salesforce and Agentforce

When AI Transformation Breaks Workforce Trust: Signal from WiseTech

AI’s Move Into Managerial Workflows: What Amazon and Accenture Reveal

Merck’s $1 Billion AI Move and the Real Shift From Pilot to Production