The recent U.S. Census Bureau data revealed an unexpected dip in AI tool usage among large enterprises. Specifically, the biweekly Business Trends and Outlook Survey (BTOS) found that the share of companies with over 250 employees using AI dropped from roughly 14% in mid-June to under 12% by late Augustitpro.com<\/a>. This marks the sharpest decline since the survey began tracking AI adoption in 2023itpro.com<\/a>. In a climate where AI has been hyped as transformative, this development has prompted questions about whether the initial \u201cAI boom\u201d is entering a cooling phase. In this report, we delve deeply into the data and its context, examine expert explanations for the slowdown, relate it to the broader AI hype cycle, compare it with alternative adoption metrics (from sources like Ramp and UBS), and assess the future outlook from enterprise, policy, and hype-cycle perspectives. All findings are backed by sources and visualized where appropriate.<\/p>\n\n\n\n AI adoption rates by firm size (1\u20134 to 250+ employees) from Nov 2023 to Aug 2025. Large firms (red line) saw a peak in mid-2025 followed by a notable dropapolloacademy.com<\/a>apolloacademy.com<\/a>.<\/em><\/p>\n\n\n\n The Business Trends and Outlook Survey (BTOS)<\/strong> is a biweekly Census Bureau survey covering ~1.2 million U.S. employer businesses (all nonfarm sectors)census.gov<\/a>tomshardware.com<\/a>. Starting in late 2022, BTOS added questions on AI, asking firms if they had used any artificial intelligence tools<\/strong> (e.g. machine learning, natural language processing, virtual agents, or voice recognition) to help produce goods or services in the last two weeks<\/strong>apolloacademy.com<\/a>. This broad definition captures AI use in core business operations (production of goods\/services), and responses are weighted to be representative of the entire U.S. economycensus.gov<\/a>. The survey\u2019s large sample and biweekly frequency make it a valuable barometer of AI adoption trends in near real-timecensus.gov<\/a>.<\/p>\n\n\n\n Timeline of Adoption Rates:<\/strong> From late 2023 through mid-2025, overall business AI adoption was steadily rising before the recent dip. As of Oct\u2013Nov 2023, only about 3.8\u20133.9%<\/strong> of U.S. businesses reported using AI in productioncensus.gov<\/a>. By mid-2024 that share had climbed above 5%<\/strong>tomshardware.com<\/a>, and continued upward into 2025 \u2013 reaching roughly 9%\u201310%<\/strong> by early summer 2025itpro.com<\/a>. Large enterprises led the charge: companies with \u2265250 employees were the most likely to use AI, with their adoption rate peaking just shy of 14% in June 2025<\/strong>itpro.com<\/a>. However, in the subsequent biweekly surveys, large-firm adoption slipped back to ~12% by late August 2025<\/strong>, a notable pullback after consistent growthitpro.com<\/a>. Other size categories saw mixed trends; many mid-size and small businesses had lower overall usage and some experienced plateaus or modest declines around the same periodtomshardware.com<\/a>apolloacademy.com<\/a>. The drop among large firms \u2013 the first significant decline since AI tracking began \u2013 suggests that even the most AI-forward companies hit a moment of hesitation in the summer of 2025.<\/p>\n\n\n\n Industry Breakdown:<\/strong> The BTOS data also show wide disparities by industry<\/strong> in AI adoption. Sectors like Information Technology and related services report the highest usage<\/strong> \u2013 well into the double digits \u2013 whereas many traditional industries remain in the low single digits. For example, in late 2023 the Information sector (software, data hosting, media, etc.) already had about 13.8% of businesses using AI<\/strong>, far above the national averagecensus.gov<\/a>. By early 2024 this sector\u2019s adoption climbed to roughly 18%<\/strong> (the highest of any sector) according to Census researchapnews.com<\/a>. Professional, Scientific and Technical Services also showed above-average AI uptake (~9% of firms by late 2023)census.gov<\/a>. In contrast, industries such as Construction, Agriculture, and Accommodation\/Food Services lagged behind \u2013 for instance, only ~1\u20132%<\/strong> of businesses in construction or hospitality were using AI as of early 2024apnews.com<\/a>. These gaps reflect that AI tools have so far been more easily adopted in data-intensive and tech-centric fields, whereas sectors with less digital infrastructure or narrower margins have seen slower uptake. Notably, firm size and industry often interact<\/strong>: larger firms (which are over-represented in tech\/finance sectors) use AI more than small firms, though an interesting finding was that the very smallest businesses (micro-firms)<\/strong> slightly outpaced mid-sized firms in some adoption metricsapnews.com<\/a> \u2013 perhaps because small startups are experimenting with AI, whereas some mid-tier firms are more cautious. Overall, the June\u2013August 2025 period showed a slight downtick<\/strong> nearly across the board, but especially for large enterprises, even as total adoption (all firms) hovered around ~9\u201310% nationallyitpro.com<\/a>. The Census Bureau cautions that the biweekly data can have some sampling variance and \u201cfluctuations in statistical accuracy,\u201d but the trend of a pullback in summer 2025 appears robusttomshardware.com<\/a>.<\/p>\n\n\n\n Methodological Notes:<\/strong> Each biweekly BTOS sample is large, but individual firms only respond once every 12 weeks (the sample is split into rotating panels)census.gov<\/a>. This means the \u201ccurrent use of AI\u201d question captures different slices of businesses over time, and the results are smoothed by the Census (e.g. Apollo\u2019s analysis applied a six-survey moving average)apolloacademy.com<\/a>. The question wording \u2013 focusing on AI use in producing goods\/services \u2013 might undercount companies using AI solely in back-office or non-production functions (more on this in Section 4)itpro.com<\/a>. Still, the overall trend<\/strong> is clear: after a surge in late 2023 and early 2024 (when AI usage nationally jumped from ~3\u20134% to ~6% then ~9%tomshardware.com<\/a>itpro.com<\/a>), the trajectory leveled off in mid-2025<\/strong>. The dip among large firms in particular has raised eyebrows, given that this cohort had previously been steadily increasing its adoption and is often seen as a bellwether for enterprise technology trendsitpro.com<\/a>. This data provides the foundation for our investigation \u2013 next, we explore why<\/strong> this slowdown might be happening.<\/p>\n\n\n\n Experts and industry observers have proposed several, often overlapping, reasons for the recent cooling in AI uptake among big companies. These range from economic considerations to organizational and technical hurdles. We have grouped the commentary into five thematic categories:<\/p>\n\n\n\n One of the most cited reasons for caution is the challenge in demonstrating a clear return on investment (ROI)<\/strong> for AI initiatives. After an initial frenzy of AI pilot projects, many firms are finding that the promised efficiency gains or revenue boosts are not materializing \u2013 at least not quickly enough. A striking statistic from an MIT report in mid-2025 found that 95% of organizations reported essentially no<\/em> measurable financial return from their AI investments<\/strong> so farmlq.ai<\/a>. In other words, only a tiny 5% of AI pilots were \u201cextracting millions in value\u201d while the vast majority had \u201cno measurable P&L impact,\u201d<\/strong> essentially a failure to deliver tangible ROImlq.ai<\/a>. Similarly, S&P Global data (cited by The Economist<\/em>) noted that the share of companies abandoning most of their generative AI pilot projects<\/strong> rose to 42% in 2025<\/strong>, up from just 17% the year beforetechcrunch.com<\/a>. This suggests a wave of disillusionment as early experiments failed to justify their costs. Nicole Kobie, writing for IT Pro<\/em>, observed that some large enterprises are \u201cgrowing frustrated at poor returns on investment\u201d from AI, which is contributing to hesitationitpro.com<\/a>. In fact, the Census data drop itself is interpreted as a sign that initial enthusiasm is giving way to ROI skepticism \u2013 companies that rushed to deploy AI are reevaluating if those deployments are actually yielding benefits.<\/p>\n\n\n\n Why is ROI proving elusive? Often AI projects incur significant up-front costs in data preparation, integration, training, and change management, while the gains (like labor savings or new revenue) are uncertain or slow. A biweekly MIT Sloan<\/em> survey described an \u201cAI paradox<\/strong>\u201d: big firms led in launching AI pilots, but few pilots scaled to full production because they didn\u2019t show quick wins<\/strong>mlq.ai<\/a>mlq.ai<\/a>. Many executives likely overestimated AI\u2019s short-term benefits. A Gartner analyst pointed out that in the real world, \u201cproper due diligence\u201d<\/em> and extensive work are needed to make AI reliable and effective, rather than expecting it to \u201cmagically\u2026everything happens\u201d<\/em> after feeding it datacio.com<\/a>. When those magic results didn\u2019t instantly appear \u2013 and in some cases AI even underperformed (e.g. producing errors or \u201challucinations\u201d) \u2013 business leaders pulled back, unwilling to invest further without a clear business case. Surveys indicate that fewer than 30% of AI leaders report their CEO is satisfied with the returns on AI spending<\/strong> so fargartner.com<\/a>. This ROI pressure is leading to more scrutiny: projects now need to prove value or risk cancellation. In summary, the hype outpaced the economic reality<\/strong>, and the recent slowdown reflects a correction as companies demand that AI prove its worth before further rollouts.<\/p>\n\n\n\n Another barrier is the cost and complexity of integrating AI systems<\/strong> into existing operations. Many organizations underestimated how difficult and expensive it is to implement AI at scale. Initial pilot programs \u2014 often using off-the-shelf models or APIs \u2014 can be set up quickly, but scaling them company-wide or embedding them into core business workflows is much harder. Gartner\u2019s Birgi Tamersoy noted that \u201cAI systems inevitably make mistakes\u201d<\/em> and achieving robustness requires building a lot of surrounding infrastructure and safeguardscio.com<\/a>. This includes data engineering, model monitoring, error handling, security measures, etc., which can be technically challenging and expensive. As generative AI models tackle more complex tasks, computing and energy costs skyrocket<\/strong>; running advanced AI can cost millions in cloud infrastructure and electricity, which directly impacts the bottom linecio.com<\/a>. For example, companies that enthusiastically deployed large language models have reported eye-opening bills for cloud GPU usage. At some point, organizations must ask if the benefit outweighs these costs<\/strong>cio.com<\/a> \u2013 and some are concluding it doesn\u2019t, leading them to scale back or optimize their AI usage.<\/p>\n\n\n\n Integration with legacy systems is another issue: enterprise IT stacks are often outdated or not designed to work seamlessly with AI tools. A 2025 analysis identified \u201cIntegration with legacy systems\u201d<\/em> as a top-7 challenge in enterprise AI adoption, noting that many companies\u2019 existing software and databases are \u201cincompatible with AI workflows\u201d<\/strong>, requiring substantial investment in new infrastructure or middlewarestack-ai.com<\/a>. Without this, AI projects remain siloed experiments. Data quality<\/strong> problems also surface \u2013 models perform poorly if fed fragmented or biased data, yet cleaning and unifying corporate data is a monumental task for large firmsstack-ai.com<\/a>. All of these integration hurdles contribute to delays and disappointments. Indeed, Apollo\u2019s chief economist Torsten Sl\u00f8k commented that the Census data likely reflects companies \u201ctapping the brakes\u201d<\/em> on AI after encountering these practical roadblocksitpro.com<\/a>. In concrete terms, some firms that raced to implement AI found themselves needing to spend \u201chuge sums on data centers\u201d and IT upgrades<\/strong> to support these systemsitpro.com<\/a>. If budgets blow out or timelines slip due to integration challenges, executives may pause further adoption until a clearer, cost-effective path emerges. Thus, high implementation costs and technical complexity<\/strong> are key factors tempering the initial rush to AI.<\/p>\n\n\n\n The human element inside firms is another decisive factor. Introducing AI often requires significant changes in workflows and can generate employee resistance or even fear. Many organizations lack a culture that fully embraces fast-paced technological change, especially one as potentially disruptive as AI. Employees may worry about job security<\/strong> (e.g. \u201cWill AI replace my role?\u201d) or simply be reluctant to trust and use AI tools in their daily work. This can lead to low adoption even when the technology is available \u2013 e.g. a company might license an AI software, but employees under-utilize it or managers don\u2019t mandate its use due to skepticism.<\/p>\n\n\n\n Analysts have indeed flagged \u201corganizational resistance\u201d<\/strong> as a top challenge: workers often \u201cfear change, don\u2019t adopt tools, or resist new AI-driven processes,\u201d<\/strong> which can seriously hinder an enterprise\u2019s AI effortsstack-ai.com<\/a>. Ingrained processes and middle-management inertia can slow down the integration of AI into business decisions. Additionally, some companies lack \u201cAI literacy\u201d<\/em> among staff and leadership, as Gartner observes \u2013 if people don\u2019t fully understand AI capabilities, they can\u2019t effectively implement or trust themgartner.com<\/a>. Training employees and redesigning roles to work alongside AI is a non-trivial task that many firms are only beginning to tackle. The initial hype often glossed over the fact that AI adoption is as much a people<\/em> challenge as a tech challenge<\/strong>. As reality sets in, organizations are recognizing that they need to invest in change management: communicating a vision for how AI will augment (not just replace) staff, providing upskilling opportunities, and creating an internal culture of experimentation rather than fear. Companies with a rigid or risk-averse culture likely put the brakes on AI projects when they encountered internal pushback or low uptake, contributing to the observed slowdown. Going forward, addressing cultural readiness is seen as key: a Thomson Reuters study found firms with a clear AI strategy (often implying better organizational alignment) were 3.5\u00d7 more likely to realize benefits<\/strong> than those without onethomsonreuters.com<\/a>thomsonreuters.com<\/a>, highlighting how important the human factor is in successful adoption.<\/p>\n\n\n\n Regulatory and ethical uncertainties<\/strong> around AI are also casting a shadow and causing some firms to hesitate. In mid-2025, the policy landscape for AI is in flux: governments are scrambling to craft rules on data privacy, AI transparency, bias, intellectual property, and liability. This uncertainty can make companies nervous about deploying AI at scale, especially in heavily regulated industries (like finance, healthcare) or for use cases involving sensitive data. Business leaders don\u2019t want to charge ahead with AI projects only to find out later that they run afoul of new regulations or expose the firm to legal risks.<\/p>\n\n\n\n For instance, concerns about data privacy and compliance<\/strong> are significant \u2013 AI often requires large datasets (which may include personal or customer data), raising questions under laws like GDPR in Europe or various U.S. privacy laws. A survey of AI adoption challenges noted that \u201cprivacy, security, and compliance\u201d<\/em> issues are a major hurdle, as AI systems can \u201craise risks around sensitive data and regulatory compliance\u201d<\/em>stack-ai.com<\/a>. Until clear guidelines are established, many companies proceed cautiously to avoid reputational or legal damage from a misuse of AI. Additionally, high-profile controversies (e.g. AI algorithms exhibiting bias or making unethical decisions) have made executives aware of the ethical governance<\/strong> needed \u2013 and not every organization is prepared to implement AI ethics boards or rigorous oversight, so some choose to slow down deployment.<\/p>\n\n\n\n On the regulatory front, concrete frameworks are emerging but not yet fully settled. In the EU<\/strong>, the forthcoming AI Act is a comprehensive regulation that will impose requirements on AI systems (e.g. transparency, human oversight, and risk management for \u201chigh-risk\u201d AI). Parts of it start taking effect in 2025, with enforcement in 2026\u20132027eversheds-sutherland.com<\/a>. European businesses (and U.S. multinationals operating in Europe) are now preparing for compliance \u2013 for example, providers of general-purpose AI models will face new obligations as of August 2025, and by 2027 even AI systems already on the market must complyeversheds-sutherland.com<\/a>. This creates a moving target for companies: some may delay certain AI deployments until they see the final regulations to ensure they don\u2019t invest in a direction that becomes non-compliant. In the United States, regulatory action has been slower, but there are executive initiatives and agency guidelines being developed. Sector-specific rules (like FDA guidance on AI in medical devices, or FTC warnings about AI in consumer protection) add to the patchwork of considerations. Additionally, the legal environment around AI-generated content (IP ownership, copyright) and AI-related liability is still being tested. All of this uncertainty imposes a \u201cwait-and-see\u201d mindset<\/strong> for more risk-averse firms. Gartner\u2019s 2025 analysis noted that \u201cgovernment regulations\u2026may impede [some] GenAI applications\u201d<\/em> and that organizations are grappling with governance challenges like bias and transparencygartner.com<\/a>. In sum, the current regulatory limbo<\/strong> makes AI a compliance risk \u2013 and many enterprises have hit pause until rules of the road are clearer.<\/p>\n\n\n\n Finally, there is a broader strategic re-evaluation<\/strong> happening in boardrooms regarding AI. The past couple of years (2022\u20132023) saw what many call an \u201cAI arms race,\u201d where companies felt pressure not to miss out on the next big thing. This led to huge investments \u2013 IT Pro<\/em> notes \u201ctrillions in AI capex\u201d<\/em> were announced or spent by large firms in a bid to ride the waveitpro.com<\/a>. But now that the initial hype is cooling, some of those aggressive bets are being reconsidered. Arpit Gupta, a finance professor at NYU, remarked upon seeing the Census adoption downturn that \u201ctrillions in AI capex should probably be reconsidered\u201d<\/strong>itpro.com<\/a>. This encapsulates a shifting sentiment: instead of \u201cAI at any cost,\u201d executives are adopting a more measured approach, integrating AI into a longer-term strategy rather than as a quick win.<\/p>\n\n\n\n In practice, strategic re-evaluation means companies are pivoting from experimentation to consolidation<\/strong>. Many are taking stock of which AI pilot projects actually showed promise and which did not. Rather than continuing to scale pilots blindly, they may shelve a number of them (as evidenced by the 42% abandonment rate mentioned earlier) and double-down on the few that align with core strategy or demonstrated ROI. The era of \u201cjust try AI everywhere\u201d is giving way to \u201clet\u2019s focus on where it truly adds value.\u201d<\/strong> For example, a bank that experimented with dozens of AI use cases might determine that only fraud detection and customer chatbots yielded positive results, and thus channel investment to those areas while cutting others.<\/p>\n\n\n\n Additionally, macroeconomic conditions cannot be ignored \u2013 2023\u20132024 were marked by rising interest rates and cost pressures, leading companies to prioritize cost-effective investments. If AI projects are not yielding near-term results, CFOs might curtail their budgets until a clearer business case is presented. This strategic pragmatism is part of normalizing the technology beyond the hype. In interviews, some tech leaders have admitted that the hype led to unrealistic expectations<\/strong>, and now there\u2019s a need to \u201creset\u201d<\/em> and educate stakeholders on what AI can and cannot do in the near termcio.com<\/a>cio.com<\/a>. The slowdown in adoption<\/strong> can thus be seen as a strategic pause<\/em> \u2013 companies taking a breath after the frenetic hype peak, refining their AI roadmaps, and ensuring that future AI deployments are tightly aligned with business goals and come with proper change management. This more cautious, strategy-driven approach contrasts with the earlier FOMO-driven rush, and is likely healthier in the long run, but it does show up in the data as a plateau or dip in adoption as the frenzy cools.<\/p>\n\n\n\n The trajectory of AI adoption in enterprises right now closely maps to what Gartner\u2019s famous Hype Cycle<\/strong> model would predict. Over the last two years, AI \u2013 especially Generative AI<\/strong> \u2013 shot up to what Gartner calls the \u201cPeak of Inflated Expectations<\/strong>,\u201d and is now sliding down into the \u201cTrough of Disillusionment<\/strong>.\u201d This conceptual framework helps contextualize the recent pullback in adoption: it\u2019s a classic case of a technology moving past the hype peak and into a phase of realism (and sometimes pessimism) before genuine productivity gains are eventually realized.<\/p>\n\n\n\n Gartner\u2019s 2025 Hype Cycle for Artificial Intelligence<\/strong> explicitly places Generative AI on the descent. According to Gartner analysts, \u201cGen AI enters the Trough of Disillusionment\u201d<\/em> in 2025 as organizations develop a more sober understanding of its capabilities and limitsgartner.com<\/a>. In 2023, generative AI (think ChatGPT and similar tools) was at the absolute peak of hype \u2013 touted as game-changing for every industry, attracting massive investments, and yielding plenty of media buzz and pilot projects. By mid-2024 and into 2025, however, the inflated expectations began to deflate<\/strong>. The reasons mirror what we discussed earlier: inconsistent results (e.g. AI chatbots that sometimes err or \u201challucinate\u201d), lack of reliability for critical tasks, and the significant work needed to maintain and refine these systems. Gartner notes that many organizations \u201chave run into problems with [GenAI\u2019s] robustness and reliability\u201d<\/em> and discovered that the \u201chype\u2026downplayed much of the work needed to reap its benefits.\u201d<\/em>cio.com<\/a>cio.com<\/a>. In essence, the shiny promise met messy reality \u2013 leading to disappointment.<\/p>\n\n\n\n We are \u201cexactly at that inflection point\u201d<\/strong> on the curve. Gartner\u2019s expert Birgi Tamersoy says the excitement in the enterprise has \u201cpassed its peak\u201d<\/em> and now we need better use cases and more accurate results to \u201crenew the enthusiasm\u201d<\/em>cio.com<\/a>. Other analysts have started invoking the term \u201cAI winter\u201d<\/strong> \u2013 referring to past periods where AI development slowed after hype cycles. For instance, by late 2025 some financial analysts warned of a potential new AI winter if investment enthusiasm wanes; Fortune magazine noted these warnings as AI stock prices wobbled, and Tom\u2019s Hardware<\/em> highlighted that narrative in coverage of AI\u2019s boom-to-chilltomshardware.com<\/a>. While an \u201cAI winter\u201d (a prolonged trough) isn\u2019t a certainty, the language shows a clear sentiment shift from unbridled optimism to caution and even pessimism in some quarters.<\/p>\n\n\n\n However, the Hype Cycle also suggests what comes next<\/strong>: after the trough of disillusionment, the technology (if fundamentally value-adding) enters a \u201cSlope of Enlightenment\u201d<\/strong> and eventually a \u201cPlateau of Productivity.\u201d<\/strong> Gartner predicts that generative AI will \u201ctake 2 to 5 years to climb out of the trough\u201d<\/em> and reach a stable, productive stagecio.com<\/a>. In practical terms, this means we can expect the current slowdown to persist in the near term \u2013 with continued skepticism and slower growth in adoption \u2013 but over the next few years, as best practices are learned and the tech matures, adoption could pick up again in a more sustainable way. Importantly, Gartner identifies trends that will aid this maturation: for example, AI engineering and ModelOps<\/strong> (model operations) are being emphasized as foundational practices to reliably and scalably deploy AIgartner.com<\/a>gartner.com<\/a>. These are the kinds of \u201cboring\u201d but necessary capabilities that signal moving past hype into real productivity. Likewise, emerging solutions to AI\u2019s current limitations (like \u201ccomposite AI\u201d that combines multiple techniques to overcome single-model weaknessescio.com<\/a>cio.com<\/a>) show that the industry is actively working through the trough.<\/p>\n\n\n\n In short, the AI hype cycle narrative<\/strong> aligns well with what we observe: 2023 saw the \u201cPeak of Inflated Expectations<\/strong>\u201d (everyone jumping on AI with perhaps unrealistic hopes), and now in 2025 we\u2019re in the \u201cTrough of Disillusionment<\/strong>\u201d (some firms pulling back, citing difficulties and unmet promises). The Census Bureau data drop among large firms is a quantitative symptom of this qualitative cycle. The saving grace is that beyond the trough, a more measured and realistic growth<\/strong> in AI adoption is likely. As one expert put it, \u201cwe have definitely not overestimated the medium- and long-term implications of LLMs [large language models]\u201d<\/em> even if we \u201coverestimated [their] potential in the near term\u201d<\/strong>*cio.com<\/a>. The task now is to climb out of the trough by solving the real challenges \u2013 something that will be discussed in the next sections on alternative data signals and future outlooks.<\/p>\n\n\n\n While the Census BTOS offers one view of AI uptake (based on self-reported usage in production), other data sources paint a nuanced picture. Notably, financial and investment analyses<\/strong> from firms like Ramp and UBS suggest that AI adoption, in some respects, continues to grow robustly \u2013 especially when measured by spending or across particular sectors. These perspectives provide a counterpoint to the narrative of a broad slowdown, indicating that the reality may differ by data source, metric, or segment of the market.<\/p>\n\n\n\n Ramp\u2019s AI Adoption Index:<\/strong> Ramp, a fintech company, uses anonymized corporate credit card and bill payment data from 40,000+ businesses to track how many firms are paying for AI products or services. This Ramp AI Index<\/strong> captures actual spending on AI tools (e.g. subscriptions to AI software, API usage charges, etc.), which is an indirect but concrete measure of adoption. According to TechCrunch reporting on Ramp\u2019s data, the proportion of businesses using (i.e. spending on) AI reached about 41% in May 2025<\/strong>techcrunch.com<\/a>. Strikingly, Ramp\u2019s methodology found much higher adoption rates than the Census survey \u2013 e.g., 49% of large businesses<\/strong> had some AI spend, 44% of mid-sized, and 37% of small companies by Maytechcrunch.com<\/a>. (This is versus Census figures in single digits overall, because Ramp counts any<\/em> AI-related usage, not just in product production. Also Ramp likely captures many instances of AI used in support functions or via third-party services that a survey respondent might not think to report.) Ramp\u2019s data thus suggests AI penetration is already quite widespread<\/strong> \u2013 over a third of U.S. businesses were paying for AI tools by early 2025ramp.com<\/a>. Moreover, spending was surging: Ramp noted AI-related expenditures among its customers grew nearly 4\u00d7 year-on-year, and that \u201cover a third of Ramp customers now pay for at least one AI tool, compared to 21% one year ago\u201d<\/em>ramp.com<\/a>. This indicates rapid growth through 2024 and early 2025.<\/p>\n\n\n\n However, even Ramp saw signs of leveling off by mid-2025<\/strong>. After ~10 months of steady increase, their index plateaued at ~41% in Maytechcrunch.com<\/a>. In other words, the share of companies adopting AI (by this measure) stopped growing for the first time in almost a year. This parallels the Census finding of a summer plateau\/decline. Ramp\u2019s economist Ara Kharazian suggested that businesses might be hitting a natural adoption ceiling in the short term, or at least a pause as companies digest their AI investments<\/strong>. The alignment is notable: both a direct usage survey (Census) and a spending-based measure (Ramp) indicate that late spring\/summer 2025 was an inflection point where explosive growth gave way to a more hesitant trend. Ramp\u2019s data also highlights a potential reason<\/em> for the pause \u2013 firms realizing \u201cthere\u2019s a limit to what today\u2019s AI can do\u201d<\/em> and encountering failures. They cite, for example, the fintech company Klarna\u2019s experience: it attempted to replace support agents with AI but had to backtrack and rehire staff due to \u201clower quality\u201d customer service outcomestechcrunch.com<\/a>.<\/strong> Many companies likely experienced similar setbacks. Additionally, 42% of companies in an S&P Global survey said they have now shelved most of their GenAI pilots<\/strong> (as noted earlier)techcrunch.com<\/a>. These real-world checks could explain why, despite more businesses than ever having tried AI (per Ramp\u2019s high adoption percentage), the growth rate<\/em> of new adopters is slowing.<\/p>\n\n\n\n UBS\u2019s Analysis:<\/strong> Meanwhile, UBS\u2019s Wealth Management research team presented an optimistic macro view in June 2025. In a report titled \u201cOverall AI adoption is far from a peak,\u201d<\/em> UBS analysts interpreted the Census data up to Q2 2025<\/strong> quite differentlyubs.com<\/a>. They noted that \u201cthe latest survey showed another step-up\u201d<\/em> in usage: national AI adoption rose to 9.2% of firms in Q2 2025<\/strong>, up from 7.4% in Q1 and 5.7% in Q4 2024ubs.com<\/a>. From UBS\u2019s perspective, this was evidence of a robust secular trend, with AI diffusion accelerating. They even compared it to the trajectory of e-commerce, noting that \u201ccrossing the 10% threshold\u201d<\/em> (which was imminent for AI) took e-commerce 24 years to achieve, whereas AI might do it in a couple of yearsubs.com<\/a>ubs.com<\/a>. UBS argued that \u201coverall AI adoption is far from a peak\u201d<\/strong>, pointing out that some industries already report 25\u201330% adoption with tangible use cases<\/strong>ubs.com<\/a>. For example, they cited real-world success stories: tech companies using AI coding assistants to save hundreds of millions of dollars, or heavy AI usage in customer service (PayPal automating ~80% of support interactions with AI)ubs.com<\/a>. These anecdotes illustrate that in certain leading firms and sectors, AI implementation continues to deepen and yield benefits.<\/p>\n\n\n\n How do we reconcile UBS\u2019s bullish view with the recent slowdown data? One key is timing and scope: UBS published their take in June 2025, using quarterly aggregated data (up to Q2) \u2013 which indeed showed strong growth from 2024 to early 2025ubs.com<\/a>. The more granular biweekly data picked up the softening in late Q2 into Q3 (June\u2013August) that a quarterly view might gloss over. It\u2019s possible that from UBS\u2019s vantage in June, the broader trend still looked clearly upward (and they projected it forward confidently), whereas by August the picture had changed modestly. Additionally, UBS and Ramp highlight that adoption is not uniform<\/strong> \u2013 even if large firms paused, smaller firms or specific industries might still be increasing their use. The Census data itself showed overall usage tick up from 8.8% to 9.7% around Augustitpro.com<\/a>, implying that while large firms\u2019 share dipped, some smaller ones may have continued to adopt, keeping the national number roughly flat-to-slightly-up. UBS\u2019s point about some industries being as high as 30% indicates pockets of very high adoption<\/strong> (information\/tech, finance, etc.) which could continue to rise even if the average across all businesses stagnates temporarily.<\/p>\n\n\n\n There\u2019s also a methodological difference<\/strong>: surveys like Census may undercount casual or indirect AI use, whereas spending data might overcount instances (e.g. a firm that spends $50 on an AI SaaS tool once is marked as an \u201cAI adopter\u201d by Ramp). Ramp itself notes its metric might miss free AI usage but also might count companies that are merely experimentingramp.com<\/a>ramp.com<\/a>. The truth likely lies between these measures. What\u2019s clear is that large enterprises (250+ employees) are the vanguard<\/strong> \u2013 they reached the highest adoption levels first and are the first to show a plateau. Smaller firms lag in percentage adopting, but that also means many have room to grow and could continue to increase adoption even as large firms regroup. In fact, Bloomberg<\/em> reported in mid-2025 that \u201cAI usage is spreading among small businesses\u201d<\/em> even as big firms temper expectationstomshardware.com<\/a> (Tom\u2019s Hardware cited a June survey of 1,500 small businesses that saw AI usage dipping among them too, but other data suggests small firms are still catching up).<\/p>\n\n\n\n Convergence or Contradiction?<\/strong> In summary, these alternative data points both converge with and diverge from the Census trend<\/strong>. They converge in suggesting that summer 2025 saw a leveling off<\/strong> after a period of rapid growth. They diverge in magnitude and framing<\/strong>: Ramp shows a much higher absolute adoption rate (41% vs ~10%) by counting any AI spend \u2013 implying many firms have at least dabbled in AI \u2013 whereas Census\u2019s stricter usage definition shows a lower core adoption. UBS remains optimistic that we haven\u2019t come close to a saturation point for AI in business, highlighting ongoing growth stories and forecasting continued exponential<\/em> investment (they estimate global AI capital expenditure will grow another 33% in 2026, reaching $480 billion after a 60% jump in 2025)ubs.com<\/a>. The Census data doesn\u2019t contradict that long-term growth \u2013 it may simply reflect a short-term pause<\/strong>. In fact, one could interpret the pause as a healthy breather; adoption curves are rarely straight lines upward. Even during the rise of cloud computing or e-commerce, there were moments of plateau before re-acceleration.<\/p>\n\n\n\n What all sources agree on is that AI adoption is highest in large companies and certain tech-forward sectors<\/strong>, and that we are still in early days relative to AI\u2019s potential. The disillusionment phase<\/strong> might be a temporary hurdle on an otherwise upward trajectory, as UBS suggests. The next section will look ahead to how enterprises and policymakers are responding, and where AI adoption might go from here, balancing the cautious signals with the continuing momentum in the background.<\/p>\n\n\n\n Despite the recent slowdown in adoption growth, AI is far from yesterday\u2019s news. The question on everyone\u2019s mind: What comes next?<\/strong> In this section, we assess the future from three angles \u2013 how enterprises plan to proceed with AI investments, how policy and regulation might shape the playing field, and where we stand on the hype cycle (and what that implies for the near future).<\/p>\n\n\n\n For enterprises, the brief chill in AI enthusiasm does not mean they are giving up on AI \u2013 rather, many are recalibrating their AI investment strategies<\/strong> to be more strategic, value-driven, and sustainable. We can expect a few key trends in how companies approach AI in the coming years:<\/p>\n\n\n\nU.S. Census BTOS Data on AI Adoption (June\u2013August)<\/h2>\n\n\n\n
Why the Slowdown? Expert Perspectives on AI Adoption Challenges<\/h2>\n\n\n\n
1. Difficulty Proving ROI<\/h3>\n\n\n\n
2. High Integration Costs and Technical Complexity<\/h3>\n\n\n\n
3. Organizational Culture and Change Resistance<\/h3>\n\n\n\n
4. Regulatory Uncertainty and Risk<\/h3>\n\n\n\n
5. Strategic Re-evaluation after the Hype<\/h3>\n\n\n\n
The AI Hype Cycle: From Peak of Inflated Expectations to the Trough of Disillusionment<\/h2>\n\n\n\n
Contrasting Data: Ramp and UBS \u2013 Is AI Adoption Still Booming or Leveling Off?<\/h2>\n\n\n\n
Future Outlook and Next Steps<\/h2>\n\n\n\n
Enterprise Strategies Going Forward<\/h3>\n\n\n\n
\n