Synthetic's $10M Gamble on a Fully Autonomous AI Accountant

SAN FRANCISCO, CA – May 14, 2026 – A San Francisco startup has secured $10 million in seed funding to build what it calls a truly autonomous AI bookkeeper, a service that employs no human accountants. The company, Synthetic, is betting that artificial intelligence can entirely replace human oversight in one of business's most critical functions, a vision that has attracted top-tier investors despite the founder's own public skepticism about the technology's current reliability.

The funding round was led by Khosla Ventures, with participation from Basis Set Ventures and a roster of high-profile operator-investors including Shopify CEO Tobi Lütke. The investment signals significant confidence in Synthetic’s audacious goal: to offer accrual-basis bookkeeping for a starting price of just $49 per month, a fraction of the cost of human-staffed services. The AI will connect directly to a company's financial systems—banks, payroll, and even inboxes—to prepare a clean set of books ready for a tax preparer.

Yet, the company’s founder and CEO, Ian Crosby, is candid about the immense challenge ahead. "I'm not sure if it's yet technologically possible to make this work," Crosby stated. "AI is notoriously unreliable, and no one wants to entrust their accounting to a system which might get it wrong... I'm not sure if that's going to be 6 months or 6 years."

A Calculated Rejection of the Human-in-the-Loop Model

Crosby's pursuit of a fully autonomous system is deeply informed by his past. In 2012, he co-founded Bench Accounting, which grew into North America's largest bookkeeping service for small businesses by pairing software with an army of human bookkeepers. Despite raising over $100 million, the hybrid "human-in-the-loop" model proved economically challenging. The company struggled with the high operational costs of maintaining a large workforce, leading to issues with service quality and high customer churn before it ultimately ceased independent operations.

This experience forged Crosby's conviction that the human-staffed model has a "structural ceiling on price, speed, and availability that no amount of operational improvement will overcome." Synthetic is the direct result of that lesson—a deliberate pivot away from augmenting human labor toward replacing it entirely. While competitors like Zeni and Botkeeper have gained traction by using AI to assist human accountants, Synthetic is making a riskier bet on full autonomy. By eliminating the human element, the company aims to create a service that is not only cheaper but also available 24/7, without vacations, backlogs, or the inconsistencies inherent in a human workforce.

The company is starting with a deliberately narrow focus on software, SaaS, and AI businesses, where accounting needs are often more standardized. This allows the team to iterate on its prototype with early design customers in a controlled environment before tackling more complex industries.

The Autonomous Accountant's Dilemma

The central challenge for Synthetic lies in overcoming the very unreliability that its CEO acknowledges. The market for AI in accounting is booming, projected to grow from under $5 billion in 2024 to over $90 billion by the early 2030s. However, this growth has been built on AI systems that assist, rather than replace, professional judgment. The risk of AI "hallucinations"—generating confident but incorrect information—is particularly acute in a field governed by strict rules and regulations.

Achieving an AI that is "more reliable than a human bookkeeper" requires surmounting significant technical and ethical hurdles. Financial accounting demands transparency and auditability, concepts at odds with the "black box" nature of many advanced AI models. Regulators and accounting bodies like the AICPA emphasize the need for explainable AI (XAI) and maintain that professional judgment cannot be fully outsourced to an algorithm. Any autonomous system must prove it can adhere to standards like the Sarbanes-Oxley Act and produce auditable, compliant financial statements without human validation.

"Our focus is on quality control and we're not going to release this until we feel confident that it's more reliable than a human bookkeeper," Crosby affirmed. This commitment to reliability over speed sets a high bar for the company's engineering team, which works in-person six days a week in San Francisco to tackle the problem.

The Khosla Thesis: Backing the 'Impossible' Founder

For lead investor Khosla Ventures, this high-risk, high-reward profile is precisely the draw. The venture capital firm has built its reputation on backing visionary founders using deep technology to disrupt massive industries, with a stated thesis of investing in AI that can replace experts across various fields. The bet on Synthetic is as much a bet on Ian Crosby as it is on the technology.

"This one’s quite simple," said Jon Chu of Khosla Ventures. "You have a large, valuable problem that will inevitably be solved by AI. A founder who’s spent multiple decades working on the problem with near perfect founder market fit. And resilience and grit that’s been forged through multiple founding experiences."

Crosby's decade-long immersion in the bookkeeping industry, including the painful lessons from Bench, gives him a unique operational understanding of the problem he's trying to solve. Investors see this "founder-market fit" as a critical de-risking factor for an otherwise technologically speculative venture. The participation of seasoned operators from companies like Shopify, Brex, and Opendoor further underscores this confidence in the founder's vision and execution capability.

Beyond Bookkeeping: The 'Business-in-a-Button' Vision

While autonomous bookkeeping is the immediate goal, Crosby's aspiration extends far beyond accounting. He envisions a future where an entrepreneur can "press a button and watch a real, running company assemble itself around their idea: the website, incorporation, bank accounts, payments, accounting, and everything else." In this future, starting a software business would be as easy as creating a new code repository.

Accounting, with its rule-based nature and vast digital data trails, serves as the perfect entry point for this grander vision of full-scale business automation. By solving the complex problem of autonomous financial record-keeping first, Synthetic aims to build the foundational layer of trust and technology required to automate the entire startup creation process. It's a long-term play that reframes the company not merely as a bookkeeping tool, but as a potential platform for the future of entrepreneurship itself.