01Real Estate Finance
AI Is Rewriting Real Estate Underwriting — and Deal Teams Still Using Spreadsheets Will Lose to Algorithms
Commercial real estate underwriting has long been a labor-intensive, analyst-heavy process: manual rent roll analysis, cap rate modeling built in Excel, and comparable sales pulled from fragmented databases. That process is being compressed to hours. A 2026 CBRE survey of 190 CRE investment teams found that firms using AI-assisted underwriting platforms cut deal underwriting time by an average of 44% and reduced human error rates in rent roll analysis by 62%. The shift is driven by machine learning models that ingest property operating statements, lease abstracts, macro rent growth data, and local vacancy signals simultaneously — generating probabilistic IRR models and sensitivity tables in real time. Platforms including Cherre, Blooma, and Lev are enabling deal teams to screen 10x more opportunities in the same analyst hours, while lenders from regional banks to CMBS shops are deploying AI models to automate initial credit memos, flag covenant risks, and benchmark loan-to-value ratios against comparable transactions. For CFOs at real estate operating companies, the implication is direct: your equity and debt partners are increasingly making capital allocation decisions through AI-augmented lenses. Financial packages that aren't structured for machine ingestion — clean data, standardized templates, consistent metrics — will score worse in algorithmic screens before a human ever reads them.
CFO TakeawayAudit your investor reporting package against what AI underwriting models prioritize: consistent NOI build-ups, clear capital expenditure schedules, and standardized operating metrics. If your package requires a human to interpret it before data can be extracted, it's already working against you in AI-driven deal screening.
02Strategic Finance
CFOs Are Replacing the Annual Budget with AI-Driven Scenario Libraries — and Boards Are Starting to Demand It
The static three-scenario forecast — base case, upside, downside — is becoming a liability for CFOs presenting to sophisticated boards. In a macro environment defined by rate volatility, geopolitical disruption, and AI-driven competitive shifts, a three-point forecast range no longer represents the uncertainty that actually governs business outcomes. A 2026 Deloitte CFO Signals survey found that 58% of large-company CFOs now run more than 20 distinct planning scenarios per quarter, up from 14% in 2024. AI-powered planning platforms are enabling this shift by automating the scenario generation process: a CFO defines key assumptions and variable ranges, and the system generates 50-100 scenario permutations, ranks them by probability and impact, and flags which combinations of assumptions produce the most significant variance to plan. Tools including Planful, Jedox, and Quantrix are allowing finance teams to build persistent scenario libraries — live models that update as actuals come in, rather than static snapshots rebuilt from scratch each quarter. For boards, the implication is a shift from asking 'what's your forecast?' to 'which scenarios are you managing for?' — a fundamentally different conversation that requires CFOs to present probabilistic ranges with supporting logic rather than a single number defended with conviction.
CFO TakeawayBuild a scenario library before your next board meeting: five pre-modeled scenarios covering your top two macro risks, each with defined triggers (the conditions that move you from one scenario to another). Boards are increasingly rewarding CFOs who show dynamic thinking over point estimates.
03Audit & Compliance
Agentic AI Is Autonomously Completing Audit Workflows — and Big Four Firms Are Deploying It Faster Than Clients Realize
The first wave of AI in audit was assistive: AI flagged anomalies, surfaced exceptions, and helped auditors sample more intelligently. The second wave — agentic AI — is structurally different. Rather than assisting auditors, autonomous agent systems are now completing entire audit sub-workflows without human intervention: reconciling GL accounts against sub-ledgers, testing journal entry completeness across full populations (not samples), cross-referencing vendor master files against payment histories to flag duplicate payments, and generating control documentation from system logs. A 2026 EY report on AI in external audit found that agentic workflows were handling 34% of traditional audit procedures autonomously at pilot clients, with human auditors focused exclusively on judgment-intensive work: risk assessment, management interviews, and sign-off. For corporate finance teams, this creates two distinct pressure points. First, internal audit departments that haven't deployed agentic tools will face increasing cost and quality pressure relative to peers. Second, the quality of data infrastructure — clean GL, properly governed vendor masters, complete transaction logs — has become the primary determinant of how much value agentic audit tools can deliver. Companies with fragmented ERP environments and manual reconciliation processes will see minimal benefit; those with unified data layers will compress audit cycles significantly.
CFO TakeawayIdentify your top three highest-effort audit preparation tasks — typically intercompany reconciliations, PBC list compilation, and control testing documentation. These are the highest-ROI entry points for agentic AI. If you're planning a finance transformation initiative in the next 18 months, data consolidation that enables agentic audit should be the first workstream funded.