Like this? There's a new one every week.
SubscribeAI is collapsing the cost of services that have risen for decades. The mechanism is quiet, the consequences will be loud, and your cost of living is about to change.
For most of your adult life, the price of nearly everything that involves a human brain has gone up. Legal fees, medical bills, tuition, accounting, customer support, software development. The pattern felt like gravity: services get more expensive every year, and you adjust.
AI breaks that pattern. When a task that required a skilled human can be performed by software at a fraction of the cost, the price of that task drops. When the cost drops far enough, services that were luxuries become utilities. And when enough services flip from luxury to utility at the same time, the economic landscape shifts beneath everyone's feet.
This is already happening. The numbers are specific, the trajectory is steep, and the categories getting hit first will surprise you.
The economics of AI-driven deflation, in plain terms.
The deflation mechanism has three stages, and each one amplifies the last.
First, AI handles the routine portion of skilled work. A customer service agent spends most of their time on repetitive queries. An AI handles those, and the cost per interaction drops from $4.60 to $1.45Industry benchmark from AllAboutAI 2025 customer service statistics, a 68% reduction in cost per customer interaction after AI implementation.. The human agent still handles complex cases, but you need fewer agents for the same volume.
Second, AI compresses the time that skilled work requires. A physician using ambient AISoftware that listens to doctor-patient conversations and automatically generates clinical notes, coding suggestions, and documentation. documentation tools sees 2.8% more patients per week while producing better notes. Translators using neural machine translation finish projects in half the time. The skill stays human, but the throughput doubles.
Third, AI enables work that previously required specialists to be done by generalists. A small business owner who once paid a lawyer $400 per hour for contract review can now use an AI legal tool that costs $50 per month. The lawyer's expertise is still valuable for complex matters, but the floor has dropped out of the routine work.
These are not projections. These are measured reductions in businesses that adopted AI tools between 2024 and 2025. Content production timelines compressed by 80%. Medical coding cycle times fell 60-80%. Contact center operational costs dropped 30%. The pattern repeats across every service category where AI can handle a meaningful share of the cognitive load.
The deflation cascade follows a predictable sequence.
Services with the highest ratio of routine cognitive work to total work get hit first. The more a job involves applying known patterns to known inputs, the faster AI compresses its cost.
Tax preparation, bookkeeping, and payroll processing sit at the front of the line. These tasks involve structured data, well-defined rules, and minimal judgment calls. AI handles them well today. Next come insurance claims processing, medical billing, and basic legal document drafting. Then customer-facing services: tech support, appointment scheduling, loan applications.
The second wave hits creative and analytical work. Marketing copy, graphic design templates, market research summaries, competitive analyses. These tasks require more judgment, but AI has reached a quality threshold where the output is good enough for most use cases. A Fortune 500 company still hires a branding agency. A local restaurant uses AI to write its menu descriptions and social media posts.
The third wave is where it gets interesting. Education, healthcare diagnostics, legal counsel, financial advising. These are high-trust services where human judgment matters and where prices have climbed relentlessly for decades. AI will lower costs here too, but the mechanism is different. Rather than replacing the professional, it multiplies them. A doctor who uses AI diagnostic tools can handle a larger patient panel without sacrificing quality. The cost per visit falls because the fixed cost of the doctor's time is spread across more patients.
How AI-driven efficiency shows up in inflation data.
Economists at Savvy Wealth estimate that AI efficiency gains are creating an annual drag on the Consumer Price Index of 0.5 to 0.7 percentage points. If sustained, that drag is enough to re-anchor long-run U.S. inflation near 1.8%.
To understand why that number matters, consider the difference between 3% inflation and 1.8% inflation compounded over twenty years. At 3%, something that costs $100 today costs $181 in two decades. At 1.8%, it costs $143. That $38 gap, multiplied across every service you pay for, represents a meaningful improvement in purchasing power for ordinary households.
AI-driven deflation is the rare economic force that helps the people who need help most. When the cost of legal advice, medical care, and education drops, the benefit accrues disproportionately to people who were priced out of those services before.
The January 2026 CPI report put headline inflation at 2.4% and core inflation at 2.5%. Within those numbers, specific categories show the AI effect. Airline fares dropped 3.4% in 2025 after AI-optimized scheduling and pricing. Gasoline fell 3.4% as AI-managed logistics reduced distribution costs. Software-as-a-service pricing has flattened across the industry as AI-powered competitors undercut incumbents.
The counterforce is energy. Data centers running AI models consume electricity at a rate that could account for nearly half of U.S. electricity demand growth through 2030. Wholesale electricity prices rose 10-18% year-over-year in regions with heavy data center construction. AI makes things cheaper, but the infrastructure that runs AI makes energy more expensive. The net effect depends on which force dominates, and so far, the efficiency gains are winning.
Deflation sounds great until you look at who was selling the thing that got cheaper.
Every service that gets cheaper represents income that someone used to earn. The customer service agent whose job was worth $4.60 per interaction now competes with software priced at $1.45. The junior copywriter whose first job was writing product descriptions finds that job automated. The bookkeeper, the paralegal, the medical coder. These roles sit directly in the deflation path.
Lower prices for legal help, medical visits, financial advice, education, and daily services. Access to professional-grade work that used to require professional-grade budgets. More purchasing power, more options, more autonomy over personal finances and health.
Pricing pressure on routine expertise. Fewer entry-level positions in knowledge work. Revenue compression for firms built on billable hours. Need to move upmarket toward complex, judgment-intensive work where AI augments rather than replaces.
History offers a partial guide. When ATMs replaced bank tellers for routine transactions, the number of bank branches actually increased because the cost per branch fell. Tellers shifted to sales and advisory roles. The banking industry employed more people after ATMs, not fewer. But the transition was bumpy, the new roles required different skills, and the people who lost teller jobs were not always the ones who landed advisory roles.
The AI deflation cycle will follow a similar arc with one critical difference: it is hitting many sectors simultaneously. ATMs affected banking. AI affects every industry that involves routine cognitive work, which is nearly all of them. The speed and breadth of the transition means that retraining and labor market adjustment will need to happen faster than it ever has before.
I spent eight years learning ICD-10 codes. Memorized thousands of them. My accuracy rate was 97%, which was top-tier in my department. The hospital paid me well because billing errors cost them millions, and I prevented those errors. That was my value.
Last March they brought in an AI coding system. It hit 96% accuracy on day one. By June it was at 98.5%. They let go of twelve coders. I stayed because I'd been there longest and because someone needed to review the AI's flagged cases. My job went from coding 200 charts a day to reviewing 40 flags.
The hospital saves about $2 million a year on coding staff now. Patient bills got processed faster. Fewer denials. I keep my salary but I can feel the ground shifting. My daughter wants to go into healthcare administration. I told her to learn something the machines are bad at. She asked me what that is. I'm still thinking about it.
A practical timeline for how AI deflation affects household spending.
The deflation is already here for digital services. It arrives for professional services within two to three years. Physical goods take longer because atoms are harder to make cheap than bits.
Software gets cheaper fastest because AI can write, test, and deploy code. The marginal cost of a software feature approaches zero when AI generates it. Professional services follow because AI can handle the research, drafting, and analysis that junior professionals currently provide. Healthcare administration deflates as billing, scheduling, and documentation become automated. Education costs fall as AI tutoring proves effective for core instruction, pushing human teachers toward mentorship and complex problem-solving.
Physical goods are the holdout. Manufacturing requires robots, materials, and energy, all of which have their own cost curves. AI helps with logistics, quality control, and design optimization, but the savings are incremental rather than transformative. Your groceries will get a little cheaper. Your legal bill will get a lot cheaper.
For the first time in a generation, the cost of expertise is falling.
If you are a consumer, the next five years will bring access to services that were previously reserved for people with larger budgets. Quality legal advice, personalized education, competent financial planning, efficient healthcare. These things will cost less because the cognitive labor that produced them will be shared between humans and machines.
If you provide those services, the same period will demand adaptation. The floor of your market is dropping, and the ceiling moves to wherever human judgment, creativity, and trust remain irreplaceable. That ceiling is higher than most people expect, but reaching it requires climbing.
New essays every week. No spam. Unsubscribe whenever you want.