AI Is Leverage. Expertise Is the Multiplier. Tony Malott · Published 2026-08-30 https://shareplane.malott.ai/artifacts/ai-is-leverage-expertise-is-the-multiplier/ SharePlane · Mini-thesis AI Is Leverage. Expertise Is the Multiplier. What becomes possible when deep experience, modern AI, and a desire to serve converge. By Tony Malott · August 29, 2026 The governing point AI compresses execution. Expertise shapes the result. Service gives the capability purpose. Preview copy. This is Semantic Candidate v02, not the final SharePlane semantic lock. By Tony Malott · August 29, 2026 What becomes possible when deep experience, modern AI, and a desire to serve converge. AI is changing the time scale of knowledge work. Things that once took days can now sometimes happen in hours, and work that previously required several specialized people can increasingly be explored by one capable person using the right tools. Writing, research, software development, analysis, design, and automation are all being compressed. That matters, but speed is the least interesting part of what is happening. The more consequential change appears when AI is combined with deep experience, judgment, context, and a clear reason for doing the work in the first place. Give two people access to exactly the same model and they can produce radically different results. The difference is not merely who writes the cleverer prompt. It is what each person knows, what they notice, what they question, and what they recognize when the machine confidently wanders into the weeds. The Model Is Only One Part of the System I have spent enough time around complex systems to be suspicious whenever somebody explains performance by pointing at one component. AI is no different. The model matters, obviously, but so does everything around it: the context it receives, the knowledge it can retrieve, the tools it can use, the boundaries placed around its authority, the way its work is validated, and whether anything learned during the process survives into the next interaction. A powerful model sitting alone in a chat window is useful. The same model inside a well-designed operating environment is a different class of capability. That surrounding machinery is becoming an engineering discipline of its own. Context has to be organized rather than repeatedly reconstructed. Knowledge has to survive beyond one conversation. Tools and data sources have to be connected safely. Workflows need defined authority, validation, recovery, and evidence of what actually happened. The model is only part of the system. The surrounding machinery determines how much useful capability can actually be brought to bear. That distinction matters because models will continue improving and, over time, access to excellent models will become less unusual. The durable advantage moves toward how intelligently those models are integrated with people, knowledge, systems, and work. The model is only part of the system. The surrounding machinery determines how much useful capability can actually be brought to bear. Experience Becomes More Valuable, Not Less One of the stranger assumptions about AI is that increasing machine capability somehow makes human expertise less important. I think the opposite happens once AI becomes capable enough. Experience contains accumulated pattern recognition. You know which questions usually reveal the real problem, which architectural shortcuts come back to collect interest later, which numbers deserve another look, which failure modes are plausible, and which beautifully written explanation is complete nonsense. Much of that knowledge was expensive to acquire because somebody, frequently you, paid tuition in the form of things going wrong. AI gives that accumulated judgment a much faster execution layer. An experienced person can explore more alternatives, investigate more deeply, prototype ideas sooner, automate repetitive work, and test assumptions that previously would not have justified the effort. The leverage comes from allowing expertise to operate across a much larger surface area without proportionally increasing the amount of manual execution required. That changes where the experienced person's value sits. Less time has to be spent manufacturing the artifact. More time can be spent deciding what should exist, whether it makes sense, what could fail, and whether the result is actually useful. Why This Matters for Small Organizations This becomes particularly interesting for churches, nonprofits, and other organizations that possess substantial human knowledge but limited technical capacity. A church, for example, may have decades of institutional knowledge distributed across staff, volunteers, committees, old documents, email, financial systems, spreadsheets, shared drives, and quite a few things that everybody assumes somebody else understands. One person knows the building. Another understands the finances. Someone remembers why a particular policy exists. Someone else knows which spreadsheet is actually current, a distinction civilization has somehow failed to automate. The problem is not that those people lack capability. The problem is that much of the organization's capability is trapped inside individuals and disconnected systems. Modern technology can help preserve that knowledge, connect it, make it understandable, and pass it from one generation of leadership to another. AI lowers the cost of doing some of that work dramatically. A small organization no longer necessarily needs a large technology department to obtain capabilities that would have been impractical only a few years ago. That is a much more interesting proposition than simply giving everybody another application to learn. The Goal Is Not More Technology I am not interested in making a church technologically sophisticated for the sake of being technologically sophisticated. Nobody gets extra credit because the committee meeting now has a more impressive software stack. Technology earns its place when it makes the underlying work better. That may mean making financial information easier for trustees to understand, preserving institutional knowledge before somebody leaves, reducing repetitive administration, making responsibilities clearer, connecting information currently scattered across several places, or helping a new leader understand why things work the way they do without conducting an archaeological expedition through old email. Sometimes the best result is simply giving people time back. That is also where I think the boundary around AI matters. A church should protect human discernment, pastoral judgment, theology, relationships, and responsibility. Those are not administrative inefficiencies waiting to be automated. But protecting those things does not require preserving every spreadsheet hunt, duplicated data entry, missing document, and manual reporting process surrounding them. AI should not write the sermon. It should give the pastor time to write it. That is the distinction I care about. Different People Bring Different Gifts Every functioning community depends on people contributing different things. Some people teach, preach, sing, organize, manage finances, maintain buildings, care for people, or quietly keep everything from falling apart behind the scenes. Technology happens to be one of the things I know. I have spent much of my career working with software, architecture, automation, infrastructure, security, data, and complex systems. AI now gives me substantially more leverage across all of those areas. I can investigate faster, build faster, connect information faster, test ideas sooner, and take on work that previously would have required far more time or people. This seems like a fairly obvious place for me to contribute. I am not particularly useful in the choir, nobody should assign me to repair the roof, and the world continues to function adequately without a Tony Malott public-speaking tour. But I can build systems. I can take complicated technology and make it understandable. I can connect information that currently lives in separate places. I can automate work that people should not have to keep doing manually, and I can help preserve knowledge that otherwise disappears when the person carrying it walks out the door. That is something useful I can give. The leverage model capability compounds when the layers reinforce one another 01 Experience + judgment Domain knowledge · scar tissue · discernment 02 AI + machinery Context · tools · workflows · validation 03 Purpose + service Useful outcomes · less friction · more human attention AI makes capability move faster. Expertise gives it direction. Service gives it purpose. The Opportunity Is Leverage The most compelling use of AI is not replacing people. It is increasing what capable people can accomplish with the knowledge they already possess. A finance volunteer can spend less time assembling numbers and more time understanding what they mean. A pastor can spend less time hunting through disconnected information and more time with people. Trustees can see the condition of an organization without rebuilding the picture from several spreadsheets. New volunteers can inherit institutional knowledge instead of beginning at zero. None of that removes the human responsibility for judgment. It moves human attention toward the parts where judgment actually matters. The same principle scales well beyond churches. Small organizations can acquire capabilities that once required significantly larger budgets, specialized teams, or both. Experienced individuals can operate across a larger domain. Experts can spend less of their time performing mechanical execution and more of it applying expertise. That is leverage. Technology is something I know. If that leverage can make the work easier for people already serving the church, that is exactly where I want to use it. Where This Is Going AI models will continue becoming more capable, less expensive, and more widely available. As that happens, simply having access to a strong model becomes less differentiating. The differentiation moves elsewhere: knowledge, context, architecture, judgment, trust, integration, and purpose. It moves toward the operating environment surrounding the model and toward the expertise of the people directing it. That is why I do not see AI primarily as a replacement story. I see it as a multiplication story. Technology is something I know. AI gives me considerably more leverage with it. If that leverage can reduce friction, preserve knowledge, strengthen an organization, and make life a little easier for people already giving their time in service to others, then that is exactly where I want to use it. AI makes capability move faster. Expertise gives it direction. Service gives it purpose. Evidence behind the thesis Check the work, not just the conclusion. Public research, authority, lineage, and author testimony are labeled separately. Sources can corroborate, challenge, or bound the argument; they do not replace Tony Malott's judgment. Portable public record Take the complete artifact with you. The deterministic package contains a self-contained offline article, the exact public-route snapshot, canonical public metadata, receipt, source text when available, plain-text context, claim ledger, source records, and a member-hash manifest. Download full artifact package Read plain-text context Inspect package manifest 0 public sources Sources, authority, and lineage Each record states the role it plays. Research support and governance provenance are not treated as interchangeable. Public boundary. The Work does not criticize church staff or volunteers and does not claim AI removes the need for expertise. It frames technology expertise as one contribution among many forms of service. 0 sources 0 governed claims 1 portable package SOURCE REFERENCES Consult the source references retained in the native presentation.