BofA Slaps Salesforce With Underperform Rating, $160 Price Target: Is the AI Story Falling Flat?

Bank of America reinstated Salesforce with an Underperform rating and a $160 price target, below the $178.50 share price and far under the $268.05 consensus target. The bear case centers on AI-driven seat-pricing compression, limited upsell, and weak Agentforce monetization, despite FY26 revenue of $41.5 billion, FY26 EPS of $12.52, and $800 million in Agentforce ARR. The stock is already down 33% year to date, and the downgrade may pressure sentiment more than fundamentals.

Analysis

This is less a single-name downgrade than a valuation reset for the entire agentic-software complex. If AI features start substituting for seats, the market should stop capitalizing CRM like a durable subscription compounder and start underwriting it more like a slower-growth workflow platform with declining pricing power; that is negative not only for CRM, but for adjacent software names selling copilots, copilots-as-a-service, and automation layers that rely on “attach” economics. The first-order damage may be limited, but the second-order effect is more important: once enterprise buyers see automation reducing labor-hours per license, procurement gets far tougher across the sector, and renewal expansion becomes harder for every vendor.

The key catalyst is not the next quarter’s headline revenue; it is whether AI usage shows up as incremental monetization or simply as a bundle that preserves retention while compressing unit economics. The market will likely tolerate weak conversion for a few quarters, but if ARR growth from AI products fails to accelerate by the next 1-2 earnings cycles, the multiple can de-rate further because the base case shifts from “future growth optionality” to “defensive cash flow.” Conversely, any evidence that AI drives net-new module adoption rather than seat replacement would force a rapid re-rating, especially given how far sentiment has already reset.

The stock is likely range-bound to lower over the next 1-3 months unless management can quantitatively prove expansion economics. The contrarian point: the consensus may be overestimating how quickly AI destroys pricing, because large enterprises often pay for workflow control, auditability, and integration, not just headcount replacement; that means the moat could persist longer than bears expect even if seat growth slows. But the burden of proof now sits with management, and absent a sharp inflection in deferred revenue mix or AI attach rates, the asymmetry still favors fading rallies rather than buying dips.

Key Decisions for Investors

• Short CRM into earnings via a 1-2 month horizon, using stock or put spreads; thesis is multiple compression if AI monetization remains qualitative rather than measured, with downside toward the low-$160s if guidance disappoints.
• Pair trade: long ORCL / short CRM over the next 1-3 months; ORCL has a more credible monetization bridge from AI infrastructure demand, while CRM faces the higher risk of seat cannibalization.
• For event-driven accounts, buy CRM downside hedges 30-60 days out and finance by selling upside calls; risk/reward favors paying for convexity because the stock can re-rate quickly on any evidence of decelerating expansion.
• Avoid adding to broad software-beta longs until next earnings season; the read-through is negative for workflow and automation names where AI is marketed as an uplift but may prove deflationary to pricing.