The AI-Native Safe-Choice Inversion

Sources#

• The ERP for the AI Revolution is here

Summary#

John Glasgow's account of the macro shift that made it possible to sell Campfire (an unknown startup) into mission-critical ERP against NetSuite: "what's safe" flipped. Historically, buying the entrenched legacy incumbent was the defensible, can't-get-fired choice. Once AI took off in finance (~end of 2024), being the incumbent came to mean not being AI-native — and boards/executives began actively pushing toward AI-native vendors. The buyer's risk calculus inverted: the new, unproven, AI-native option became the one with executive "air cover."

The inversion#

"Buying the legacy version was considered very safe. But once AI started to take off… being the incumbent meant you were not AI-native. There was a flipping of the narrative — the board and executives were saying we want AI-native."

The consequence Glasgow names: even when the line-level accountant "wasn't fully ready to embrace AI," they had a blessing from above to buy something new "that nobody had heard of and their auditor wasn't familiar with yet." Top-down demand for AI-native gave bottom-up buyers cover to take the risk. "What's safe is now actually the opposite answer."

Why ERP specifically was primed (the "why now")#

Glasgow's deeper thesis on timing: every layer of the finance tech stack turned over in the last 5–10 years — payroll, spend management (the Brexes of the world) — except the core general ledger. A finance person would log into a slick modern spend-management tool and then hit the ancient ERP; the contrast was "very acute." That gap created latent demand ("all finance software doesn't need to suck") that buyers began seeking out — and AI-native was the banner the demand organized under.

Why this is counter-positioning, not just timing#

This is a textbook counter-positioning instance for Seven Powers Applied to AI: incumbents can't easily flip to AI-native without cannibalizing their existing business, retraining their org, and rebuilding on a foundation model — so the startup's positioning is one the incumbent is structurally reluctant to copy. The inversion isn't just a tailwind; it's a durable wedge as long as "AI-native" remains something the incumbent can't credibly claim. (Glasgow says Campfire is "consistently told we have the best AI… maybe because we're the only one with our own foundation model.")

Generalizable pattern#

The shape transfers beyond ERP: in any mission-critical, slow-moving category, AI can invert the buyer's default. The questions to test for a new category:

1. Is the category mission-critical (so "safe" historically meant the incumbent)?
2. Did adjacent layers already modernize, making the incumbent's staleness acute and visible?
3. Is there top-down pressure to be AI-native that gives risk-averse buyers air cover?
4. Can the incumbent credibly claim AI-native — or is it structurally counter-positioned out of doing so?

Connections#

• John Glasgow / Campfire — the source case
• Seven Powers Applied to AI — counter-positioning is the Power this inversion exploits; AI erodes incumbents' switching-cost/process advantages
• Narrow Wedge into a Legacy Market — the inversion is the tailwind; the narrow wedge is how you actually land the first customers
• Product Velocity as Moat — "best AI / shipping every day" is what sustains the AI-native claim over time
• AI-Native Startup Lifecycle — the Founder's Playbook's macro story, told from the buyer's risk-perception side
• Compounding Data Moat — once switched, the AI-native vendor accrues the data/workflow lock-in incumbents lack
• Printing Press Software Democratization — same era-shift framing from the supply side (who can build) vs. this demand side (what buyers consider safe)

Open Questions#

• The inversion is a one-time repricing of "safe." Once several AI-native ERPs exist, does "safe" re-stabilize around the largest AI-native vendor — and does Campfire's "we're now the largest of the new cohort" claim reflect a land-grab for that position?
• How long until incumbents bolt on credible AI and neutralize the counter-positioning — and does the custom-foundation-model claim actually defend against that?

Sources#

• The ERP for the AI Revolution is here