Target India head says retailer weighing AI tool costs amid shift to usage-based pricing

Target is re-evaluating its AI rollout as vendors such as Anthropic and OpenAI shift to usage-based token pricing, raising costs for enterprise customers. The company is still making significant investments in employee tools and plans to add $2 billion of spending this year on new stores, remodels and AI initiatives. The article signals cautious cost management rather than a material near-term financial shock.

Analysis

The important second-order effect here is not that TGT is “using more AI,” but that AI is moving from a marginal productivity enhancer to a budget-constrained operating expense. Token-based pricing creates variable cost exposure that scales with usage intensity, which favors firms with disciplined internal governance and high-repetition workflows, and hurts large enterprises that were banking on broad deployment without clear unit economics. For TGT, the likely outcome is a narrower set of high-ROI use cases in merchandising, forecasting, and labor planning rather than employee-wide adoption.

That tightening is mildly negative near term because retail is already in a margin repair phase: when management has to choose between store investment, price competitiveness, and AI tooling, AI usually loses unless it is directly tied to shrink, conversion, or inventory turns. The bigger winner could be software vendors with measurable ROI stacks rather than frontier model providers; enterprise buyers will increasingly favor workflow-specific applications that can justify spend in basis points of margin lift, not vague productivity claims. This should also pressure internal IT teams across retail/consumer to delay rollouts, which slows the second wave of enterprise AI monetization.

For TGT, the catalyst path is asymmetric: if AI spend shows up as incremental opex before revenue re-acceleration, the market will treat it as another layer of cost inflation, and the stock likely underperforms over the next 1-2 quarters. The contrarian case is that management is signaling discipline, not retreat: by forcing ROI thresholds now, TGT may avoid the common enterprise trap of overpaying for broad AI licenses that never scale. If they can use analytics to improve inventory and promo execution, the payoff could emerge over 12-18 months in gross margin and working capital rather than headline revenue.

Key Decisions for Investors

• Remain underweight TGT into the next 1-2 quarters if AI spend is framed as employee enablement rather than hard ROI; the risk/reward is still tilted negative because incremental opex arrives before measurable sales lift.
• Pair trade: long GOOGL or MSFT, short TGT for 3-6 months. Thesis: hyperscalers monetize AI infrastructure and enterprise tooling regardless of retail budget restraint, while TGT absorbs a cost line with uncertain payback.
• Consider a tactical short-dated put spread on TGT around the next earnings cycle if management commentary emphasizes broader AI rollout; downside is limited by valuation support, but a re-rating lower is plausible if margins compress.
• Look for a long opportunity in vertical software beneficiaries with quantifiable ROI in retail operations over the next 6-12 months; these names should capture budget share as enterprises abandon generic model subscriptions.
• If TGT explicitly ties AI to inventory turns or shrink reduction in future updates, cover shorts quickly and reassess for a 12-month long only on evidence of margin expansion, not AI adoption rhetoric.