Spreadsheet and data solutions provider Sourcetable launched a “self-driving” spreadsheet that allows users to simply tell the spreadsheet what they want done through natural language commands.

Sourcetable developed the solution as a way to bring advanced spreadsheet functionality to people who might struggle with basic functions like VLOOKUP or creating a pivot table. The “self-driving” autopilot capabilities give the AI complete write access and edit control to complete multi-step operations.

“AI is the biggest platform shift since the browser, with a bigger opportunity for disruption,” said Sourcetable CEO and co-founder Eoin McMillan. “Sourcetable is building the AI spreadsheet for the next billion users, be they human or AI. As AI makes analysis easier, everybody will become an analyst. Sourcetable’s AI automation ushers in a new era of productivity and human cognition.”

Summit Art Creations – stock.adobe.com

Sourcetable’s autopilot mode can complete a wide range of complex tasks, including creating and editing financial models, generating spreadsheet templates, building pivot tables, cleaning data, creating charts and graphs, editing formatting, enriching data and analyzing entire workbooks. The AI can understand data context without requiring users to pre-select ranges, interpret multiple ranges across different tabs, work with messy data, and seek human clarification when instructions are unclear.

The AI is capable of accessing anything that is publicly available on the internet, McMillan said in an email, and it can also extract data from URLs if instructed to do so. This includes Federal Reserve Economic Data, stock ticks and trading data, Yahoo finance, futures, geopolitics, market sentiment, macroeconomic analyses, Wikipedia data and much more. “There’s even a full fund manager Easter egg included in this release,” he added. 

This ability to access tools outside itself also means that users, via a virtual machine with hundreds of libraries and AI tools available, can ask the autopilot to find a more advanced tool to serve their needs by requesting the system to “download data” or “use Python” to solve a task. McMillan said Sourcetable plans to make this feature more user-friendly in the future as the technology ultimately moves toward becoming a full agentic platform and operating system. 

To discourage the AI from providing false information, the solution is built around a code-driven evaluation loop developed internally that verifies AI response in real time. Without this foundation, according to Sourcetable, self-driving spreadsheet automation would be too slow and unreliable to be trusted. McMillan said the company uses a combination of techniques to optimize results while minimizing latency. First, there’s AI-driven process supervision of inputs, outputs and prompts, effectively AI watching AI. This is combined with a code-driven audit of quantitative outputs (e.g., Python, SQL and spreadsheet output evaluation) and, finally, thought-driven techniques (e.g., Chain of Thought Reasoning and Deliberate Reasoning) to drive better results, particularly for multi-step processes. 

The new solution uses not one but many models to deliver results. While certain companies are locked into their own proprietary AI models, Sourcetable’s AI selects the optimal model for each task–including OpenAI, Anthropic, Groq, Meta (Llama), Nvidia, Prior Labs, DeepSeek and Hugging Face—and even combines multiple models for better results. McMillan explained that different models are better suited to different tasks and run better on different kinds of hardware. For example, he noted, Claude is currently best at coding, TabPFN at interpreting tabular data, Groq at fast inference, etc. Sourcetable’s AI knows model specifications and strengths, so i’s able to understand what a user is trying to do and find the best tool for it. 

While accessing public models can sometimes come with a per-prompt cost, McMillan said the company has established relationships with many service providers to ensure high rate limits and the ability to handle a large number of requests. He added that, right now, Sourcetable use a combination of manual and automated controls to prevent abuse of the system that could conceivably create large fees, though he believes the long-term cost curve indicates that AI will essentially become free, with the price of software being more aligned with value than cost of goods sold.

Prior to this release, Sourcetable did offer an AI copilot similar to many in the market that was more for formula assistance, charting and answering questions, according to McMillan. This was initially included as a SQL assistant to retrieve database data to help users who didn’t know how to write SQL, and this is how the company learned that users really wanted to use the AI for their regular spreadsheet workflows, leading Sourcetable to develop this current solution. 

“Ironically, solving the database retrieval problems forced us to build our own Chain of Thought equivalent before OpenAI released theirs publicly,” said McMillan. “That taught us how to leverage processes like CoT for multistep processes and automation, and this gave us a big head start once we shifted gears toward full spreadsheet automation via AI. Today’s autopilot moves us from answering questions to thinking and agency. It’s a big leap forward.”

Sourcetable offers both a free tier and a pro tier, which costs $20 a month. All Sourcetable users get the first two weeks free on the Pro tier and can continue using the system on a rate-limited free tier. All the regular spreadsheet and charting features are free and unrestricted. McMillan added that Pro users are Sourcetable’s revenue source. Free tier users generate no revenue, he said, “although happy users spread the word, which is the best form of marketing.”

The Internal Revenue Service’s Criminal Investigation unit has embarked on a new initiative for engaging with financial institutions as it makes greater use of banking data to uncover tax and financial fraud. 

IRS-CI released FY24 Bank Secrecy Act metrics Friday, demonstrating how it uses BSA data to investigate financial crimes. During fiscal years 2022 through 2024, 87.3% of IRS-CI’s criminal investigations recommended for prosecution had a primary subject with a related BSA filing, and adjudicated cases led to a 97.3% conviction rate, with defendants receiving average prison sentences of 37 months. IRS-CI also leveraged BSA data to identify $21.1 billion in fraud linked to tax and financial crimes, seize $8.2 billion in assets tied to criminal activity, and obtain $1.4 billion in restitution for crime victims.

Under the BSA, which Congress passed in 1970, financial institutions use suspicious activity reports to notify the federal government when they see instances of potential money laundering or tax evasion. The SARs data is used by agencies like IRS-CI to probe money laundering and related financial crimes.

A new IRS-CI initiative known as CI-FIRST (Feedback in Response to Strategic Threats) aims to establish ongoing engagement with financial institutions. They will receive quantifiable results from IRS-CI on how the agency uses suspicious activity reports to investigate federal crimes. 

“Public-private partnerships thrive when everyone mutually benefits, and to enhance our partnership with the financial industry, we plan to launch CI-FIRST which will promote information-sharing, streamline processes and demonstrate how valuable BSA data is to criminal investigations,” said IRS-CI Chief Guy Ficco in a statement.

As part of the CI-FIRST program, IRS-CI plans to streamline subpoena requests and share pointers with financial institutions on what to include in suspicious activity reports to maximize their impact. The program will address what’s working and what can be improved, offering continuous lines of communication between partners. IRS-CI headquarters will work with larger financial institutions that have a national and international presence, while its field office personnel will work with regional and community banks and credit unions.

IRS-CI special agents ran an average of 966,900 searches each year against currency transaction reports during the last three fiscal years. Close to 1,600 cases were opened in FY24 with at least one currency transaction report on the primary subject. The data also shows that 67.4% of cases opened by IRS-CI had a subject with one or more currency transaction reports below $40,000, with 50% of currency transaction reports involving amounts less than $22,230.

BSA data has also proven to be effective in helping IRS-CI combat narcotics trafficking and pandemic-era tax fraud. Since FY20, IRS-CI used BSA data to initiate nearly 1,300 investigations with ties to fentanyl and investigate alleged employee retention credit fraud totaling $5.5 billion.

As the expiration of key Tax Cuts and Jobs Act provisions looms, tax professionals are preparing for what could be another period of upheaval.

In 2017, when the TCJA was first enacted, tax departments struggled to keep pace with new regulations and guidance. According to our recent Bloomberg Tax survey of 434 tax professionals, 92% of tax professionals working in tax at the time reported that the TCJA’s implementation was moderately to highly disruptive, and 60% said it took a year or more to fully implement the changes. 

The coming year could bring more of the same. Eight in 10 respondents are moderately or very concerned about the potential impact of these changes. Yet many rely on outdated, manual processes that make adjusting quickly to major legislative changes difficult.

With the benefit of hindsight, tax professionals have a unique opportunity to apply the lessons of 2017 and invest in automation now to avoid repeating the same costly mistakes.

Manual processes still dominate tax departments

One of the most striking findings from our survey is that many tax professionals continue to rely on manual workflows despite the increasing complexity of tax compliance. Seventy-six percent of respondents said they still use Excel for tax calculations, and 63% manually gather data from enterprise risk management and general ledger systems to perform tax calculations.

These outdated processes create inefficiencies and make it harder for tax teams to respond quickly to legislative changes.

In its time, the TCJA was the most sweeping tax code overhaul in decades. It required tax departments to significantly modify or even replace their workpapers to reflect the changes. 

While 62% of survey respondents believe they can update their existing workpapers without major difficulty, one in four anticipate significant challenges, and 10% will need to create entirely new workpapers.

This manual burden could put firms at a disadvantage when deadlines are tight and compliance requirements shift rapidly.

Scenario modeling is challenging yet critical

When big changes are on the horizon, running multiple tax planning scenarios helps organizations make decisions and manage risk. Automated tax solutions streamline this process by allowing tax teams to evaluate different legislative outcomes and come up with strategies to address them.

Firms that lack automation in their tax workflows may have a tough time keeping up with the pace of change — especially if Congress waits until the eleventh hour to pass legislation, as was the case in 2017.

Eighty-eight percent of respondents reported it is moderately or very difficult to conduct scenario modeling for TCJA changes, and only half have started the process. One respondent noted, “We need as much lead time as possible to make changes to our models, and significant changes take even more time to incorporate. Running multiple scenarios is a very manual and difficult process.”

Quantifying the cost of inaction

Failing to invest in automation before a substantial tax law change can be a costly mistake.

Among respondents, 71% who experienced the enactment of TCJA in 2017 reported wishing they had invested earlier in tax technology to better manage the complexity of compliance updates. Manual processes not only slow response times but also drive costs, as nearly 40% of respondents anticipate a $100,000 or higher increase in consulting budgets if significant TCJA-related changes occur. 

By leveraging tax automation tools and centralized tax-focused software, firms can optimize how they engage with external consultants. Automation allows tax departments to take ownership of routine processes, such as calculations and compliance adjustments, reducing reliance on consultants for these tasks. Instead, consultants can be utilized more effectively on high-impact projects that drive strategic value, such as tax planning, risk management or navigating complex regulatory changes. This shift enables firms to streamline compliance while ensuring external expertise is directed toward creating lasting organizational benefits.

Preparation now means greater confidence going into 2026

The data is clear: firms investing in automation today will be better positioned to handle the upcoming tax changes confidently. Here’s how to get ahead:

• Integrate tax technology. Replace manual calculations in Excel with automated tax workpapers that integrate with source data and automate data gathering and calculation processes.
• Adopt scenario modeling tools. Invest in software that allows for real-time legislative modeling so you can analyze multiple potential outcomes before changes take effect.
• Reduce reliance on external consultants. Implement in-house tax software to keep control over your data, reduce consulting budgets and respond quickly to regulatory shifts.

With less than a year until TCJA provisions are set to expire, the time to act is now. Taking proactive steps to automate and modernize your workflows will put you in a far stronger position than companies that wait until the last minute. 

Major tax law changes can be disruptive, but with the right technology, you don’t have to relive the turmoil of 2017. Embrace tax-focused automation to remain agile, efficient and ready to navigate whatever changes come next.