Trend Synthetic content at scale is eroding the authenticity of work and workforce data
Summary
Deloitte's second 2026 trend warns that AI is blurring what is real in data about people and work. Generative and agentic systems now produce content in enormous volumes, which can obscure who created what, magnify bias and trap people in echo chambers. Organisations are losing trust in workforce data just when they depend on it most, and only 5 percent say they are making great progress on the problem. The erosion shows up in authenticity, such as embellished resumes and fake candidates, in agency, where it becomes unclear who did the work, and in critical judgment, as workers lean too heavily on AI. Deloitte urges a shift from cybersecurity to disinformation security. It also calls for helping workers judge what is real.
Classification
Evidence 1
- 2026 Global Human Capital Trends: From tensions to tipping points - Choosing the human advantage Deloitte Insights (Deloitte Development LLC) page=18;section=Fact or fabrication? AI is blurring the line when it comes to people and work 2026-01 accessed 2026-07-26
Futures articles 1
Observed signals 10
- Signal41% have used AI to automate part of their job, often unknown to the employer
- Signal42% of executives are already concerned about AI dependence for cognitive tasks
- Signal48% of executives fear AI is injecting misinformation into their own datasets
- Signal61% call data trustworthiness important, 5% report great progress
- Signal86% of top-ranking Google pages remain human-written
- Signal95% of executives doubt candidate skills data while a third of workers admit AI embellishment
- SignalA deepfake CFO on a video call induced a $25 million payout
- SignalFour in 10 organizations posted jobs with no intention to hire in 2024
- SignalGartner projects one in four job seekers could be artificial by 2028
- SignalMore than half of new web articles were primarily AI-generated as of May 2025
Part of issues 3
- IssueFrom cybersecurity to disinformation security: a new digital trust pact1 trends · 3 signals
- IssueTipping point 1: from human plus machine to human times machine4 trends · 0 signals
- IssueWorkslop: AI widens skill gaps and contaminates the data models later learn from1 trends · 1 signals
Relation types: constitutes · supports
Public id: fm-ddd44d4bffba
