The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →No. “Pacing” in AI policy concerns the speed and conditions of AI progress; business adoption is a separate question about whether and how organizations use AI. A proposal to slow or condition some frontier development does not, by itself, show that companies are adopting AI more slowly.
What does “AI pacing” mean?
The AI Policy Institute uses “pacing” to describe allowing AI progress to continue while establishing mechanisms to slow its rate if it becomes too fast. That is the Institute’s policy framing, not a universal technical definition, and proposals using the term need not share identical mechanisms or thresholds. AI Policy Institute
That policy question is distinct from measuring business adoption. A government or organization might seek safeguards around certain kinds of AI development or deployment while firms continue to adopt available tools. Whether safeguards add friction in a particular case depends on their design and implementation; the evidence cited here does not establish a universal causal effect on adoption speed.
Is AI adoption actually slowing down?
There is no useful timeless answer without specifying whose expectations, which businesses, what period, and what counts as AI use. A U.S. Bureau of Economic Analysis paper comparing expectations with reported use in the Census Bureau’s Business Trends and Outlook Survey from 2023 to 2026 describes an initial period when adoption was slower than expected, a short period when it was faster than expected, and a more recent period closer to expectations. That changing pattern is more precise than saying adoption is simply “slow.” BEA, “AI Expectations and Outcomes”
Recommended Free Tools
#1 Best Overall
Adoption figures also differ according to their population, date, definition, and denominator. For example, a Census Bureau working paper published in September 2023 used data from the 2018 Annual Business Survey and counted five technologies: automated-guided vehicles, machine learning, machine vision, natural language processing, and voice recognition. It found fewer than 6% of firms used any of those measured technologies, but adoption was just over 18% when weighted by employment. The second figure reflects workers’ exposure through their employers, not the share of firms using AI. These historical measures predate newer generative-AI survey measures and should not be treated as a current rate. Census Bureau, “AI Adoption in America: Who, What, and Where”
A newer Census Bureau working paper, published in April 2026, reports that 18% of firms used AI in a business function during the November 2025–January 2026 survey reference period; the employment-weighted figure was 32%. It also reports that 22% expected to adopt AI within six months. Those are different measures: reported current use and anticipated near-term adoption. Because the older study and newer study use different survey designs and definitions, they do not form a clean trend line. Census Bureau, “The Microstructure of AI Diffusion”
Why “a company uses AI” can hide shallow or uneven use
Firm-level adoption is only one layer. The 2026 Census study separates whether a firm uses AI, how many business functions use it, and whether workers use AI for particular tasks. Those measures do not always coincide: workers may use AI for tasks even without formal firm adoption, while some firms reporting formal adoption do not report worker task use.
Even among firms that had adopted AI in that study, 57% used it in three or fewer business functions. A firm can therefore count as an adopter without AI being integrated broadly across its operations. In a June 2026 UK government plan for the Digital and Technologies sector, report author Katie Gallagher OBE writes that “depth of integration, not headline adoption, drives productivity.” This is the plan’s position, not a universal causal law. UK Department for Science, Innovation and Technology, “AI Adoption Plan: Digital and Technologies”
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Adoption should not be mistaken for an outcome. A firm’s reported use does not by itself show that AI raised productivity, increased revenue, or changed employment. The BEA paper finds some links between stated motivations for using AI and changes in production processes, but describes the relationship between motivations and outcomes as murky. BEA, “AI Expectations and Outcomes”
Can governance and adoption happen at the same time?
Yes. Governance describes how AI is overseen; it does not automatically mean adoption stops. The U.S. Government Accountability Office’s accountability framework organizes practices around governance, data, performance, and monitoring. It identifies responsibilities and oversight challenges, but does not establish that accountability work necessarily slows deployment. U.S. GAO, “Artificial Intelligence: An Accountability Framework”
Rank #4
Likewise, Australia’s policy for responsible AI use in government explicitly aims to enable accelerated and sustainable adoption by agencies, with a framework that can evolve as technology and governance maturity change. That is evidence of an adoption-supporting policy aim, not proof that the policy has made adoption faster. Australian Government, “Policy for the Responsible Use of AI in Government”
Policy Horizons Canada’s 2025 foresight report raises a related concern: technological development may outpace decision-makers. That is a policy framing about governance capacity, not a measured comparison of business adoption rates. Policy Horizons Canada, “Foresight on AI: Policy Considerations”
How to interpret an AI adoption claim
Before comparing claims that adoption is fast, slow, rising, or falling, check what each one actually measures:
Quick Recap
- Population and geography: U.S. firms, UK businesses, and public agencies are different populations.
- Period: A report’s publication date may differ from the survey reference period or the vintage of the data.
- Definition: The technologies or uses counted as AI can change between surveys.
- Denominator: A firm-weighted rate and an employment-weighted rate answer different questions.
- Layer: Firm use, integration across business functions, and worker use for specific tasks are not interchangeable.
- Outcome: Adoption alone does not establish productivity, revenue, or employment effects.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
