What is AI co-tenancy clause negotiation? AI co-tenancy clause negotiation is the use of AI tools like Claude Opus 5.5, ChatGPT GPT-6, and Google Gemini to price a tenant's proposed co-tenancy language against your center's rent roll and loan covenants before the lease is signed, then generate counter-language that caps the downside. Most co-tenancy work in commercial real estate happens too late, after the clause is executed and the analyst is simply discovering what was agreed to. The leverage sits in the forty minutes before signature, while the rent reduction formula, the cure period, and the sunset are still negotiable. For the broader framework, see our guide to AI deal analysis real estate scoring.
Key Takeaways
- Set your co-tenancy rent reduction cap off your loan's DSCR covenant, not a market rule of thumb. The covenant defines the maximum NOI loss you can actually absorb.
- A temporary rent abatement is a cash flow and IRR problem. An exercised termination right is a capitalized value problem. Negotiate each differently.
- Six terms are genuinely negotiable: the named tenant list, the occupancy test definition, the cure period, the reduction formula, the termination gate, and the sunset.
- Cure periods longer than your covenant test window keep a triggered abatement from landing inside the same reporting quarter.
Why Negotiating Co-Tenancy Differs From Reviewing It
Reviewing co-tenancy means reading what you already own. Negotiating it means deciding what you are willing to own. Our guide to AI retail CRE due diligence covers the abstraction side: pulling every opening, ongoing, replacement, and permanent provision out of a seller's rent roll and mapping the cascade. That work is diagnostic and happens after the clauses exist. Tenant mix strategy, in our analysis of Claude retail tenant mix and co-tenancy strategy, tells you which tenants you want. This article is about what you concede to get them.
Pre-signature negotiation inverts the exercise. You start from your center's current stack of co-tenancy obligations, add the proposed one, and ask a narrower question: does this incremental clause push my worst case past a line I care about? Those lines are concrete: the DSCR covenant in your loan documents, the cash management trigger in your lockbox provisions, and the exit value you underwrote. Leverage also favors landlords right now. Retail availability held at 4.9% in Q2 2026 with average asking rents at $24.79 per square foot, up 2.4% year over year, according to CBRE Research, and JLL's 2026 US retail thematic outlook found 64% of investors plan to increase retail acquisitions while only 48% expect to sell more. Leverage you do not use at signature is leverage you pay for later.
The Six Co-Tenancy Terms Worth Negotiating
A co-tenancy clause is not one term. It is six, and tenants rarely resist all six equally. Asking AI to isolate each parameter shows which concession is cheap and which is expensive.
- The named tenant list: A clause naming three specific retailers is far more dangerous than one requiring a grocery operator of a stated size. Named brands go bankrupt; categories do not.
- The occupancy test definition: Measure on gross leasable area rather than tenant count, exclude the protected tenant's own space from the denominator, and count dark-but-paying tenants as occupied. These three choices routinely move the trigger point by 5 to 8 percentage points.
- The cure period: The window to replace the anchor before any remedy attaches. The most valuable term, and the one tenants most often concede.
- The rent reduction formula: Cap the abatement as a percentage of base rent only, not the full gross obligation, and keep common area maintenance and taxes outside it.
- The termination gate: Allow termination only after the cure period lapses, on written notice inside a short use-it-or-lose-it window, and only if the tenant is current on rent.
- The sunset: Have the clause expire after year five or six, or once the tenant recoups its tenant improvement allowance. A right that survives into a renewal term permanently encumbers your exit.
Set the Rent Reduction Cap Off Your DSCR Covenant
The right rent reduction cap is not 25% because 25% is customary. It is whatever percentage keeps you inside your debt service coverage ratio covenant if the clause triggers.
Take a 120,000 square foot neighborhood center: a 45,000 square foot grocery anchor plus 75,000 square feet of inline space across 22 tenants. Nine inline tenants, totaling 31,000 square feet at an average $28.00 per square foot, hold ongoing co-tenancy rights tied to the grocer. That is $868,000 of base rent exposed. Center NOI is $1,450,000 against $1,000,000 of annual debt service, so DSCR is $1,450,000 divided by $1,000,000, or 1.45x. The covenant requires 1.25x.
Now work backward. Holding 1.25x requires NOI of at least $1,250,000, so you can absorb $200,000 of NOI loss. As a share of the $868,000 exposed, that is 23.0%. Any cap above 23% puts you in breach if the anchor goes dark and the clause fires. A customary 25% cap costs $217,000, dropping NOI to $1,233,000 and DSCR to 1.23x, a breach. A 15% cap costs $130,200, leaves NOI at $1,319,800, and holds DSCR at 1.32x. You are no longer haggling over a customary number, you are showing the tenant where your lender's line sits. Our guide to AI CRE loan covenant monitoring covers tracking these thresholds after signing.
One nuance changes how you model the result. A temporary abatement during a cure window is a cash flow event: it pressures DSCR and lowers IRR, but it does not permanently impair NOI and should not be capitalized into exit value. An exercised termination right is lost gross leasable area, re-tenanting cost, and downtime, and it does flow through to value. At a 7.0% cap rate, a permanent $434,000 NOI loss is roughly $6.2 million of value, since value equals NOI divided by cap rate. Conflating them overstates abatement risk and understates termination risk.
Match the Cure Period to Your Covenant Test Window
Cure period length is usually negotiated as a calendar question: 9 months, 12 months, 18 months. Treat it as a covenant timing question. If your loan tests DSCR quarterly on a trailing twelve month basis, a 12 month cure period means the clause can trigger, abate rent, and surface in a covenant test before re-tenanting is finished. A longer cure period does more than buy leasing time; it keeps the abatement out of the reporting period. That matters most where springing lockbox provisions apply, because an abatement that trips a cash management trigger can divert cash flow even when the anchor replacement is already in the pipeline. See our guide to AI CRE cash management and springing lockboxes. Moving from 12 months to 18 is often free to the tenant and valuable to you.
The AI Workflow for Pre-Signature Co-Tenancy Review
This workflow runs roughly 40 minutes per proposed lease once inputs are assembled, against a half day of manual analysis.
- Build a clause library, then a house standard. Export the co-tenancy language from every executed lease into one document; Yardi, RealPage, and CoStar can supply the lease data. Have the AI draft your position on each of the six terms above and flag where your own portfolio is inconsistent. Most landlords find they conceded very different terms to comparable tenants.
- Score the incoming redline. Give the model the tenant's proposal and your house standard, and ask for a term by term gap table separating economic deviations from cosmetic ones.
- Run the cumulative cascade. Provide the rent roll with all existing co-tenancy obligations plus the proposed one, and ask for total exposed base rent and resulting DSCR under the trigger scenario.
- Generate counter-language, then route it to counsel. Ask for alternatives at your target cap, cure period, and sunset, each with a one line rationale your leasing agent can deliver verbally. AI drafts a position; it does not render a legal opinion.
For CRE investors who want this built as a repeatable process rather than a one-off analysis, The AI Consulting Network specializes in exactly this kind of workflow design.
Where AI Falls Short on Co-Tenancy
Replacement anchor standards are frequently ambiguous, using phrases like "a tenant of comparable quality and drawing power," and AI will offer a confident reading of language that has been litigated to different outcomes in different jurisdictions. Treat those readings as hypotheses for counsel. Models are also poor at judging which concession closes the deal: a gap table tells you what a term costs, not that this tenant's real objection is the signage package. Pair the output with our analysis of AI retail lease strategy and rental rate optimization and your leasing agent's read of the counterparty.
The larger risk is data quality. Cumulative exposure is only as good as your rent roll, so if three older leases were abstracted incompletely, the cascade understates exposure and the cap you negotiate will be too loose. Only 5% of organizations report achieving most of their AI program goals, and incomplete source data drives much of that gap. Investors who want help closing it can reach out to Avi Hacker, J.D. at The AI Consulting Network.
Frequently Asked Questions
Q: What is a reasonable co-tenancy rent reduction cap for a landlord to accept?
A: There is no universal number. Calculate the NOI loss that keeps you above your DSCR covenant, then express it as a percentage of the base rent exposed to co-tenancy clauses across the center. In the example above that figure was 23.0%, which makes a customary 25% cap a breach and a 15% cap safe.
Q: Can AI actually redline a co-tenancy clause, or just summarize it?
A: Current frontier models including Claude Opus 5.5 and ChatGPT GPT-6 will produce usable alternative language when given your house standard and the tenant's proposal as context. The output is a drafting starting point that still requires counsel review, not a finished legal instrument.
Q: Should the occupancy threshold be measured by tenant count or square footage?
A: Gross leasable area, and exclude the protected tenant's own space from the denominator. Counting tenants lets several small vacancies trigger a clause that a square footage test would not, and the definitional choice can shift the trigger point by 5 to 8 percentage points.