The arrival of Claude for Small Business earlier this week marked an interesting moment—and a savvy strategic move—for Anthropic. Rather than saddling web browsers with more AI slop or trying to slather AI onto perfectly good user interfaces that don’t need improving, Anthropic is attempting something both less flashy and potentially more fruitful: finding a practical, agentic AI-powered application for everyday business owners looking to make ends meet.
The bag of tricks included in Claude for Small Business is somewhat predictable, running the gamut from “ready-to-run” agentic workflows to connectors for PayPal, QuickBooks, HubSpot, Canva, DocuSign, and more. With these tools, business owners can use Claude to help to plan their payrolls, reconcile their books, analyze their cash flow, spin up promotional campaigns, and so forth. That’s all well and good, but it also entails trusting Claude to perform those bookkeeping and promotional duties accurately and thoughtfully. Many small business owners will (quite rightly) balk at the prospect of handing their tried-and-true workflows to an unpredictable AI model, even one as powerful as Claude.
Small business users were equally hesitant about computers in general at the dawn of the PC age. Sure, an Apple II or a Commodore 64 could balance checkbooks and track inventories, but not much better (or faster) than a human could. Why bother coughing up $1,500 (in unadjusted 1979 dollars) for an Apple II that wasn’t much better at bookkeeping than a person with an old-school ledger? Then as now, what was missing was a killer app—a game-changing application that takes a familiar task and transmutes it into something simple, elegant, and (in retrospect) seemingly inevitable.
The Historical Parallel: VisiCalc and the Personal Computer Revolution
In 1979, Dan Bricklin and Bob Frankston released VisiCalc, the first electronic spreadsheet for the Apple II. It turned the personal computer from a hobbyist toy into a serious business tool. With VisiCalc, a small business owner could not only track expenses and revenue but also forecast them—simply by changing a single number in a cell. The application demonstrated the core value of the PC: instant recalculation, what-if scenarios, and visual presentation of data. Suddenly, the $1,500 price tag for an Apple II looked like a bargain. VisiCalc was the killer app that drove the adoption of personal computers in offices around the world. It was later eclipsed by Lotus 1-2-3, and then Microsoft Excel, but the pattern was set.
Today’s AI landscape mirrors that pre-VisiCalc moment. Large language models like ChatGPT, Claude, and Gemini are powerful but aimless. They can generate poetry, code, and business memos, but they lack a single, universally compelling use case that makes them indispensable for the average person. Sure, Claude Code is a killer app for developers—a narrow but vocal niche—but for the rest of us, AI remains an imperfect fit. It is like trying to use a socket wrench to slice a wedding cake: it can do the job after a fashion, but it’s not the right tool.
Anthropic’s Strategic Pivot: AI for Small Business
With Claude for Small Business and Claude Cowork, Anthropic is nibbling around the edges of what VisiCalc accomplished: finding a truly useful and unique application for AI that offers tangible value to small business owners—and, by extension, to everyday users everywhere. But trying to shoehorn the agentic AI abilities of Claude Code into the world of small business may be a dead end. What makes AI terrific at crafting code—its endless creativity—is what makes it so worrisome when it comes to business. Yes, AI can build meticulously crafted spreadsheets and beautifully crafted bar charts in seconds, but they are useless if the data behind them cannot be trusted. AI models are known to hallucinate, to be overconfident in erroneous outputs, and to behave unpredictably when faced with ambiguous inputs. These traits are unacceptable when dealing with payroll, tax filings, or customer data.
Public skepticism toward AI remains high. In a recent survey, over half of respondents expressed distrust in AI-generated financial advice. The term “AI slop” has entered the lexicon to describe low-quality, automated content that pollutes web search results and social media feeds. Even when AI gets it right, the perception of unreliability lingers. The industry still searches for something that provides “calm technology”—a tool that works quietly in the background, delivering value without requiring constant vigilance or correction.
The Search for AI’s Killer App
What could that killer app be? Several candidates have emerged: AI-powered inbox organizers that triage email with high accuracy; voice assistants that truly understand context and can hold a natural conversation; automated legal document review; personalized tutoring systems that adapt to each student’s pace. Yet none has achieved the mass adoption of VisiCalc. The challenge is that a true killer app must be both simple and profound—it must solve a real problem for a broad audience in a way that feels obvious after the fact. For the PC, the problem was “I need to do calculations quickly and see how changes affect results.” For AI, the equivalent might be “I need to make sense of a huge amount of information without spending hours reading and analyzing.”
Some argue that the killer app is already here but unrecognized. ChatGPT reached 100 million users faster than any previous product. Yet widespread use does not equal deep integration. Many users treat AI as a toy or a research assistant, not as an indispensable workhorse. The difference between a tool and a toy is whether it is used daily in a central workflow. VisiCalc was used daily by accountants, managers, and small business owners. AI, for now, is used sporadically for tasks like drafting emails, generating ideas, or summarizing articles. It is not yet embedded into the fabric of our work and lives.
From Slop to Substance: Harnessing AI’s Strengths
The key is harnessing AI’s power in a different way, applying its strengths to the right applications while turning its flaws into virtues. For instance, AI’s runaway creativity—which causes it to invent facts—can be an asset for creative brainstorming, for generating alternative scenarios, or for exploring possibilities that a human might never consider. The challenge is to design interfaces and workflows that constrain the AI appropriately, so that its creativity is expressed within safe boundaries. This is precisely what enterprise AI platforms like those from OpenAI and Anthropic are trying to do with “structured outputs” and “function calling.” They give the model a sandbox to play in, with clear rules and guardrails.
Another promising direction is the “80/20 prompt” mentioned in the original article. Derived from the Pareto principle, this technique asks the AI to identify the 20% of concepts that yield 80% of the benefit. It is a crash course in any subject, perfect for learning something fast. Such prompts illustrate that AI’s value may lie not in full automation but in accelerating human learning and decision-making. A teacher can use AI to generate lesson plans, a doctor to summarize a patient’s history, a marketer to draft A/B test copy. These are not tasks that replace humans but ones that augment them.
The Road Ahead: What AI Can Learn from the PC Era
The PC revolution didn’t happen overnight. After VisiCalc came word processors, databases, desktop publishing, and eventually the internet. Each new application built on the previous one. The key was that the platform (the PC) was open, affordable, and capable of running many different programs. Today’s AI platforms are similarly open: developers can build on top of models like GPT-4 and Claude 3 with APIs, fine-tuning, and retrieval-augmented generation (RAG). The barrier to entry is low, and innovation is happening across countless startups and internal teams. Somewhere out there, a developer might be building the VisiCalc of AI—a tool that turns the raw power of large language models into a focused, reliable, and indispensable application.
For now, the search continues. Anthropic’s Claude for Small Business is a step in the right direction, targeting a specific audience with concrete workflows. But to truly become the new VisiCalc, it must overcome the trust barrier and prove that AI can handle mission-critical tasks without constant supervision. The industry also needs a better user experience—one that hides the complexity of underlying models and presents a simple, intuitive interface. The original Apple II had no hard drive, no mouse, and only 4KB of RAM. Yet VisiCalc made that clunky machine sing. Similarly, today’s AI, for all its sophistication, still feels like a promising but unfinished instrument. The killer app will be the one that makes us forget we are using AI at all—it will just feel like the right way to get things done.
In the meantime, everyday users can experiment with tools like Claude for Small Business, ChatGPT for business, or Google’s upcoming “always-on” assistant, code-named Spark. But until a true VisiCalc moment arrives, AI will remain a powerful but peripheral tool—full of potential, but waiting for its breakthrough. The lesson from history is that such breakthroughs often come from unexpected directions. Maybe it will be a spreadsheet-style interface for data analysis, or a financial forecasting tool that uses natural language to generate interactive models. Maybe it will be an AI that automatically reconciles bank transactions and flags anomalies better than any human bookkeeper. Whatever it is, it will combine simplicity, trust, and immediate value. That is the holy grail.
Until then, the industry will keep trying different approaches: bringing AI to business via integrated workflows, to education via tutoring systems, to healthcare via diagnostic aids. Each attempt adds to the collective learning. The first AI killer app may not be a single product but an ecosystem of specialized tools that together make AI indispensable. Or it could be one simple app that sweeps the world overnight. Either way, the principle remains: find the task that is difficult, repetitive, and critical for many people, and make it effortlessly easy with AI. That was VisiCalc’s magic, and that is what AI still needs.
Source: PCWorld News