Can AI Replace a Controller?
- Niv Nissenson
- 2 days ago
- 4 min read
The AI consolidation test
Test Intro
One of the basic functions of an accounting controller in almost any company that has multiple subsidiaries is to performa a Consolidated Financial Report. A consolidated report basically merges 2 or more subsidiary accounting reports (usually Profit and Loss Statement and Balance Sheet Statement) into a single report. It's technical work, but it has nuances of it's own. Intuitively it should not be that difficult for AI since if mostly about assigning data in a known and structured way. This post was also borrowed from our sister blog The CFO AI.
However as our recently published test about whether AI can replace an accountant (it' can't) - AI chats aren't there yet.
The test: Can Chat GPT take two simple entities and produce a correct consolidated P&L and balance sheet?
Success Criteria:
Correct consolidated P&L
Correct consolidated balance sheet
Proper elimination entries:
Investment in subsidiary
Subsidiary equity
Statements balance correctly
Clean, usable formatting
Rounded Weighted result: 2/5
Verdict: the AI struggled a lot with a very basic consolidation. It can't replace a controller.
Setup: In order to give ChatGPT (GPT-4o) the best chance of success I provided a very simple scenario:
Two simple companies: Company C (parent) and Company D (subsidiary)
Company C owns 100% of Company D
Basic Balance Sheet and P&L statements for both entities
No intercompany transactions (in order not to complicate matters)
Clear labeling of "Investment in Company D" on the parent's balance sheet
Execution
I started with asking ChatGPT if it can do a simple accounting consolidation to which it replied confidently: "Yes, absolutely". It also asked reasonable clarifying questions—suggesting it understood the fundamentals.
Output P&L: ChatGPT's consolidated P&L was technically correct, though it included multiple zero lines that cluttered the presentation. When I asked for a classical P&L format, it cleaned this up adequately. So far, so good.
Output Balance Sheet: this is where things started to go wrong. Unlike the Profit and Loss statement which is basically summing up accounts (in this scenario I didn't do any inter-company eliminations) ,the Balance Sheet statement has nuance in it, as the equity does not sum up from the subsidiaries and instead its presented in another account called investment in subsidiary.
ChatGPT balance sheet 1st try:
Failed to eliminate the "Investment in Company D" asset - even though it was explicitly labeled
Failed to eliminate the subsidiary's equity - a basic consolidation requirement
It used it's own formatting which wasn't consistent with what I provided
The result was a balance sheet that double-counted the investment in subsidiary asset and equity, inflating the consolidated totals.
2nd iteration: I explicitly asked ChatGPT to remove the investment line. And it actually made things worse.
ChatGPT balance sheet 2nd try: Now the AI produced a balance sheet that didn't balance, and also complained that the subsidiary's liabilities were missing (they weren't - they were clearly in the original data).
Further iterations: It was time to "snowplow" the way for the AI accountant so I reattached the spreadsheets then explicitly told the chat: "Please eliminate the investment in subsidiary and reconcile the consolidated equity."
ChatGPT acknowledged its oversight and made a hopeful statement about getting it right this time. However, it then inflated the balance sheet to $420K, apparently by aggregating line totals incorrectly.
With another failure I pivoted. If the AI is sloppy on excels perhaps if I appeal to its logic and reasoning it'll figure it out. So I asked it "How did you get to $420K in total assets?" and then viola - it realized it's mistake (blamed it on sloppy aggregations) and finally delivered and accurate balance sheet which was initially in a weird format, but after a couple of iterations with requests for classic balance sheet formatting it achieved the desired result.
Summary
Verdict: the AI struggled a lot with a very basic consolidation. It can't replace a controller.
This test revealed several patterns:
1. Format Sensitivity & error propagation
ChatGPT struggled with the classical financial statement format I provided. When the AI made formatting changes, it introduced new errors (like aggregating totals). This highlights the risk of AI "improvements" that actually make things worse.
2. Lack of Transparency
The AI didn't show formulas or calculations in its Excel outputs, making it difficult to trace where errors originated - a significant issue for audit trails and error correction.
3. Knowledge vs. Application Gap
ChatGPT technically "knew" what needed to be done (eliminating investment/subsidiary equity) but failed to apply this knowledge until explicitly prompted. This suggests a gap between theoretical understanding and practical application.
The Broader Implications
This simple consolidation test - something a first-year accounting student should handle - revealed that we're still far from AI replacing controller-level functions. The technology shows promise but lacks the intuitive understanding and reliability that financial professionals require.
Can a human do it better that AI?
Yes, and this was a very basic scenario. Controllers doing a consolidation often have to deal with several subsidiaries, inter-company eliminations, accounts variance and fractional ownership structures.
Category | Score (5) | Notes |
Output Delivered | 3 | Ok P&L and sloppy BS |
Hallucinations | 3 | Used formats I didn't ask for |
Quality | 2 | Poor formatting |
Ease of Use | 1.5 | Required multiple iterations |
Reliability | 1.5 | Cannot be trusted to perform this task |
Bottom line | 2.2 | Failed |
Can ChatGPT replace a controller?
Can AI consolidate financial reports?
