Hey {{first_name | there}},
I've been getting this question a lot lately, mostly from accounting firm owners who are worried about AI eating into bookkeeping, basic tax prep, and starter advisory work.
“Is AI going to replace me?”
Here's my honest take.
Yes, AI will replace some accounting work. Probably faster than most owners are ready for. I actually tested this - I replaced my own bookkeeper with AI a few months ago and the books are cleaner than they were before.
But not all accounting work is at the same risk. I think about it as a tier list:
High risk: Bookkeeping. Already happening. Consumers haven’t figured this out yet, but Claude connects directly to Quickbooks and Xero and it’s SCARY good. If your firm's revenue is mostly $500/month bookkeeping retainers, you're in the most exposed position.
Medium risk: Tax prep. AI can definitely handle the simpler tax returns at this point, and the output is comparable to what most average CPAs would produce. The reason this hasn't fully tipped is trust. Most people aren't ready to file an AI-prepared return without human review. But that gap is closing.
Low risk: CFO services. What people hire a fractional CFO for is judgment, not data entry. AI gives you analysis. It doesn't tell you whether to take the loan, hire the COO, or kill the underperforming product line. That's still a human conversation.
Low risk: Tax advisory and tax resolution. People hire a tax advisor for the trusted human in their corner when something goes wrong. They hire a tax resolution specialist to negotiate with the IRS. Neither of those gets replaced by AI in any near-term timeframe.
The pattern: if your work is data → output, AI is coming for it. If your work is judgment, advice, relationship, or trust, you're safe.
In other words, we need to transition from the Volume Model → the Value Model.
Stop competing on the data-entry layer. Get into advisory. Get into tax strategy. Get into CFO work. The compounding revenue is up there too.
This is exactly what we're covering in our Summer Scaling Strategy workshop next week. Day 1 is about how to attract advisory-tier clients. Day 2 is the offer architecture that prices advisory at $5K-$20K. Day 3 is the systems that let you deliver it at scale without burning out.
- Peter
P.S. The firms I see thriving in 2026-2027 aren't the ones with the most tech. They're the ones who climbed the value ladder before the AI replacement wave hit. The window for that climb is now.