Summary
The transition from AI tools to autonomous agents isn’t just a feature upgrade; it’s a structural shift in enterprise economics. We are moving from chatbots to systems that plan, execute, and iterate across workflows. This shift promises to expand the addressable work market significantly—potentially exceeding the labor cost it replaces—but it also triggers immediate labor market friction. Recent data suggests layoffs are largely a correction of COVID-era overhiring, yet the signal is clear: AI is reshaping job descriptions and displacing low-margin tasks. The unit economics of agentic work are complex; while inference costs are dropping, the volume of calls is rising. Executives must look past the hype to understand the real trade-off: automation at the margin versus the creation of entirely new value streams. The future of work isn’t about replacement; it’s about the division of labor between human intent and machine execution.
Research
The economic narrative of AI is bifurcating. On one side, the ‘efficiency’ narrative drives layoffs and job description tightening. On the other, the ‘augmentation’ narrative suggests agents will unlock work humans never performed. The data suggests a hybrid reality: genuine displacement is starting at the margins (customer service, translation), but the bulk of the current workforce reduction is a correction of the 2020-2022 hiring boom. The key differentiator isn’t just cost, but capability. Agents can now perform multi-step workflows that previously required human oversight. This expands the total addressable market for AI spend, potentially making agent-performed work more expensive than human labor in volume, despite lower per-unit costs. The critical metric for executives isn’t just ‘how much cheaper is AI?’ but ‘how much more work can it do?‘
Books
From Deerfield Green’s library of long-form research — books written to give practitioners the economic models, case studies, and strategic depth that whitepapers and blog posts can’t. Here’s what’s relevant this week.
The Next Five Years: Agentic Value Creation
BCG’s research quantifies the trajectory: AI agents currently contribute approximately 17% of enterprise AI value creation. By 2028, that figure is projected to reach 29%. Gartner adds another dimension: by the end of 2026, 40% of enterprise applications will integrate agentic capabilities. These are not chatbots with better prompts. Agents are AI systems that can plan multi-step tasks, execute actions across multiple systems, make decisions within defined parameters, and learn from outcomes. The shift from interactive tools to autonomous agents represents a fundamental change in how enterprises derive value from AI. We are moving from static outputs to dynamic workflows that operate with a degree of autonomy previously reserved for human operators.
Source: books/enterprise-ai-economics/chapters/ch18-the-next-five-years.md
Token Economics and Infrastructure Costs
Agentic workloads — where the model autonomously plans, executes multi-step tasks, uses tools, and iterates on results — multiply inference costs further. An agentic workflow might involve an initial planning step where the model decomposes a task into subtasks (1 model call), execution of each subtask involving tool calls, web searches, database queries, or code execution (3-10 model calls), and intermediate evaluation steps where the model assesses progress and adjusts its approach. This multi-layered execution increases the total token consumption significantly, creating a complex economic equation where the cost per task is higher than a simple chat interaction. The infrastructure costs are not linear; they compound as the complexity of the workflow grows.
Source: books/enterprise-ai-economics/chapters/ch03-token-economics-and-infrastructure.md
Articles
Curated from recent reporting and analysis across the industry. These are the pieces we think cut through the noise.
AI Job Displacement Statistics (2026 Trends)
By 2030, AI could automate up to 30% of U.S. jobs according to National University estimates. This isn’t just theoretical; 13.7% of U.S. workers say they have already lost a job to robots or AI-driven automation. The displacement is not uniform. It is concentrated in sectors with routine tasks and standardized outputs. While the technology is advancing rapidly, the human response is equally dynamic, with many workers actively upskilling to work alongside these new systems rather than being replaced by them. The impact varies significantly by industry, with administrative and technical roles facing the highest exposure rates.
Source: https://www.designrush.com/agency/ai-companies/trends/ai-job-displacement-statistics
How Will AI Affect the Global Workforce?
Goldman Sachs Research estimates that unemployment will increase by half a percentage point during the AI transition period as displaced workers seek new positions. The concern is that AI will lead to widespread labor displacement, but the reality is nuanced. The impact varies by occupation and the elasticity of consumer demand in that sector. As AI lowers production costs, prices may drop, increasing demand for goods and services, which in turn could offset some of the displacement effects. The transition period is expected to be turbulent, requiring significant workforce realignment and a rethinking of traditional career paths.
Source: https://www.goldmansachs.com/insights/articles/how-will-ai-affect-the-global-workforce
White Papers
Deerfield Green publishes original research on the forces reshaping labor markets, token economics, and enterprise adoption curves. These excerpts are drawn from that ongoing work.
AI Layoffs vs. COVID Overhiring
Tech layoffs attributed to AI are largely a correction of COVID-era overhiring (2020-2022), with ‘AI efficiency’ serving as narrative cover for restructuring that was inevitable regardless of AI advances. Genuine AI displacement is beginning at the margins — particularly in customer service and content translation — but accounts for a small fraction of total workforce reductions. The data shows that while headlines scream about AI replacing jobs, the underlying trend is a normalization of hiring post-pandemic. The narrative serves a dual purpose: it signals to investors that the company is future-proofing, while masking the simple fact that many tech companies simply hired too many people during the boom.
Source: whitepapers/ai-layoffs-vs-covid-overhiring/research.md
Job Description Changes Since 2020
Job descriptions are a leading indicator of how AI is reshaping workforce expectations. Since 2020, AI-related requirements have surged 134% in postings while total postings grew only 6%, revealing a structural shift in what employers value. This divergence — accelerating since ChatGPT’s November 2022 launch — is creating a new floor for skills requirements. Employers are no longer just looking for generalists; they are looking for individuals who can leverage AI tools to achieve specific outcomes. The narrowing of skill requirements suggests a move away from generalist roles toward specialized, AI-augmented roles.
Source: whitepapers/job-description-changes/research.md
Market Signals and CEO Sentiment
AI-related stocks accounted for ~75% of S&P 500 returns since ChatGPT’s launch. A workforce reduction framed around AI adoption ‘sends a signal to investors that a straightforward cost-cutting announcement does not.’ 79% of U.S. CEOs fear losing their jobs within two years if they don’t deliver ‘measurable, AI-driven business gains.’ Block’s stock surged 24% after its AI-framed 40% workforce cut. Meta stock climbed on reports of its planned 20% cut tied to AI spending. These market movements indicate that investors view AI integration not just as a cost-saving measure, but as a survival imperative.
Source: whitepapers/ai-layoffs-vs-covid-overhiring/research.md
Prototypes
We don’t just write about the future — we build it. Deerfield Green’s prototype lab produces interactive tools that let you stress-test ideas against real data. Here’s what applies to this week’s topic.
Labor Market Prediction Market
This prototype visualizes displacement risk across different sectors using observed exposure data. By aggregating theoretical LLM capability with real-world task data, the model generates a probability distribution for job loss over the next 24 months. The interface allows executives to simulate the impact of automation on specific verticals, adjusting for variables like wage rates and consumer demand elasticity. It serves as a stress test for workforce planning, moving beyond static spreadsheets to dynamic scenario modeling.
Source: prototypes/labor-market-prediction/README.md
Frameworks
From Deerfield Green’s library of strategic frameworks — structured models for measuring AI value, planning workforce transitions, and sizing transformation initiatives. These are the lenses we use internally, published so you can use them too.
Value Dynamics and Inference Costs
This framework maps the trade-off between the cost of inference and the value generated by autonomous tasks. It highlights that as inference costs drop (driven by model optimization), the volume of tasks an agent can handle increases non-linearly. The model demonstrates that in many cases, the total cost of agent-performed work will exceed human labor costs not because the per-task cost is high, but because the volume of work expands. This creates a ‘value expansion’ paradox where automation becomes economically viable only after it reaches a certain scale of deployment.
Source: frameworks/ai-value-dynamics/README.md
What’s Next
The agent revolution isn’t coming in a single dramatic moment. It’s arriving one automated workflow at a time, in the gap between what’s too simple to need a human and what’s too complex to fully automate.
References
- [1] The Next Five Years: Agentic Value Creation, Enterprise AI Economics
- [2] Token Economics and Infrastructure Costs, Enterprise AI Economics
- [3] AI Job Displacement Statistics (2026 Trends), DesignRush