Do big companies and small ones feel pressure to adopt AI differently, and move at different speeds?

This explores whether company size changes how firms respond to pressure to adopt AI. The corpus doesn't compare small and large firms directly. Several notes, though, point the same way: AI adoption doesn't spread evenly. It builds up wherever there is already capacity to use it. The clearest evidence comes from OpenAI's own usage data on ChatGPT Enterprise, which shows adoption concentrated in larger, R&D-heavy companies Who adopts enterprise AI first and how do they use it?. So on the question of who moves first, size seems to matter.

The more interesting finding is that early advantages seem to compound. A study of firms hiring from online freelance marketplaces found that some firms replace those workers with AI faster and more cheaply than others Do firms substitute labor for AI at different rates?. That study sorts firms by how exposed their work is to AI, not by headcount, so it isn't strictly about size. Its mechanism still matters here: once a firm builds internal know-how with AI, each further step gets easier. That's a scale effect, not a technology that every firm absorbs at the same pace. Anthropic's Economic Index finds a similar pattern between countries Does AI adoption follow wealth and mature over time?. Wealthier places use AI for a wider range of tasks, and their use shifts over time from handing off whole tasks to working alongside the AI. If firms behave like countries, larger firms may not just adopt sooner. They may also reach the collaborative stage sooner, while smaller firms stay focused on one narrow use for longer.

The bottlenecks may also differ by size. Benedict Evans argues that making AI tools easier to build doesn't fix the real barriers Does easier tool-building actually solve enterprise adoption problems?. Workers often don't see their own tasks as automatable, and rolling AI out across an enterprise takes decisions that span departments and long timelines. That second barrier falls mostly on large organizations. A small firm may lack the money and in-house expertise, but it can decide quickly. A large firm has the expertise but has to get through its own coordination overhead. Size also shapes how much control individual workers keep. A study of software engineers found that company policies, such as required tools, approved-tool lists and data rules, set the limits of how they use AI agents before personal preference comes into play Does personal preference shape how engineers use AI tools?. Bigger, more formal organizations probably have more of this rule-setting.

Firms of any size should be careful about what they're responding to. A survey of executives found that perceived productivity gains from AI are larger than measured ones, and that the effects concentrate in high-skill services and finance Do AI productivity gains feel larger than they actually measure?. Part of what feels like competitive pressure may be belief running ahead of results. Even inside large firms, use is uneven: marketing staff and early-career workers use the tools far more than executives Who adopts enterprise AI first and how do they use it?. Separately, people expect to be judged as less competent for using AI and tend to hide it Do people fear judgment when they use AI at work?. So company-level adoption numbers may undercount how much quiet, unofficial use is happening.

The takeaway: in this corpus, the size gap is less about access to the tools than about compounding capability. Firms that already have R&D depth and internal AI know-how get cheaper with each further step. Firms without them risk falling further behind. A direct study of small businesses would fill a clear gap in the collection.