AI Doesn’t Just Increase Buying, It Increases Conviction: Three Findings from the Q2 2026 US Consumer Survey

AI-assisted shopping has crossed the threshold from emerging behavior into material consumer pattern. Among US consumers who already use AI tools for shopping — recommendation engines, conversational shopping assistants, AI-powered search, generative product discovery — buying behavior diverges materially from the non-AI consumer baseline. The divergence is not subtle. Locus’s Q2 2026 US Consumer Survey finds AI users are more than twice as likely than non-AI users across multiple buying behaviors simultaneously — and the behaviors don’t all point in the same direction.

The headline reading is straightforward: AI increases buying. AI users try more new brands. AI users put more items in carts. AI assistants function as recommenders, and the recommendations are landing. For brand teams worried about AI commoditizing discovery, the data confirms the concern — consumers using AI tools encounter more brands more often, expanding consideration sets in ways that erode incumbent defensibility.

The reading underneath is more interesting: AI also increases conviction. Among consumers who use AI tools, more than a third (34%) say they feel confident buying fewer items — more than three times the rate of non-AI consumers (11%). The finding complicates the simple amplification narrative. AI isn’t only making people buy more; it’s also giving some shoppers the confidence to buy less, more deliberately, with stronger conviction in the choices they do make. Same tool, opposite directions, depending on the use case.

Three behaviors happening at once is the architectural reality. Wider consideration sets, larger baskets when buying, and more confident deliberation when not. The operational implications cascade across demand planning (volume volatility), last-mile (package mix), ecommerce (customer experience), and brand strategy (defensibility). For Chief Supply Chain Officers, CMOs, Heads of Ecommerce, Heads of Demand Planning, and Heads of Last-Mile evaluating 2026 architecture, the question is whether the organization reads AI behavior as three simultaneous patterns — or as a single trend that misses the structural shift.

This is a strategic look at three connected findings from Locus’s Q2 2026 US Consumer Survey, and what they mean for operational and commercial architecture.

Finding 1: AI Users Are More Than Twice as Likely to Try New Brands

The data. 18% of all US consumers say they’re likely to try new brands or products in their next purchase cycle. Among consumers who use AI tools for shopping, the figure rises to 39% — more than 2x the baseline rate.

What it means. AI-assisted discovery widens the consideration set materially. Consumers using AI tools encounter brands they wouldn’t have considered through search-and-browse discovery patterns. The expanded consideration set affects brand defensibility for incumbents and creates a discovery window for challenger brands. The shift is large enough — twice the baseline rate — that it should be material to brand and commercial strategy rather than treated as edge-case behavior.

Operational implications cascade. Demand planning models trained on incumbent loyalty assumptions may understate volume shifts as consumers test new brands at higher rates. Last-mile operations may see expanded carrier mix complexity as consumers buy from a wider brand set — which often means more shipping origins, more carrier handoffs, and more fulfillment patterns to orchestrate. Customer service complexity rises as customers interact with more brands per quarter. Brand teams face genuine defensibility erosion that traditional incumbent advantages (search ranking, repeat purchase loyalty) don’t fully address against AI-mediated discovery.

Finding 2: AI Users Put More Than Twice the Items in Their Carts

The data. 17% of all US consumers say they buy more items per order than they did previously. Among AI users, the figure rises to 37% — also more than 2x.

What it means. AI assistants function as recommenders during the shopping flow, and the recommendations are landing. Consumers using AI tools add more items per order — bundled recommendations, complementary products, accessories, alternatives all surface during the AI-assisted shopping experience in ways that lift basket size. Discovery friction drops; basket size rises. The pattern matches what AI shopping experiences are designed to produce, but the magnitude (2x baseline rate) is larger than incremental.

Operational implications cascade. Demand planning needs basket composition awareness, not just volume forecasting. The same total order volume distributed across larger baskets produces different fulfillment economics than across more, smaller orders. Last-mile package mix shifts as average order contents grow — different package sizes, different vehicle utilization, different handling patterns. Warehouse picking complexity changes as multi-item orders become more common. The shift affects unit economics throughout the fulfillment chain, not just at the customer-facing checkout.

Finding 3: AI Users Are More Than Three Times More Confident Buying Fewer Items

The data. 11% of all US consumers say they feel confident buying fewer items than before. Among AI users, the figure rises to 34% — more than 3x. This is the finding that complicates the rest of the picture.

What it means. AI isn’t only amplifying buying. Some consumers use AI tools to make more deliberate decisions, comparing options more rigorously, reaching higher confidence in fewer purchases. The same AI shopping experience that produces wider consideration and larger baskets for some shoppers produces tighter, more confident, smaller purchasing for others. Same tool, opposite directions, depending on use case. The 3x gap is the largest of the three findings — and the most consequential because it shifts the underlying narrative.

Operational implications cascade. Returns patterns evolve as deliberate buying becomes more common — deliberate purchases typically return at lower rates than impulse purchases. Demand planning must handle two opposing patterns simultaneously: AI users buying more (Finding 2) and AI users buying more deliberately (Finding 3). Customer experience architecture must serve consumers using AI for amplification AND consumers using AI for deliberation — the two patterns may overlap in the same customer at different purchase moments. Brand teams need messaging that supports both consideration breadth and decision confidence rather than optimizing for one mode.

Three Behaviors at Once: The Architectural Reading

The three findings appear contradictory only if AI behavior is read as a single trend. Read as three behaviors happening simultaneously, the findings form a coherent picture: AI mediates more of the shopping experience, and the mediation can produce wider consideration, larger baskets, or more deliberate purchasing depending on what the shopper is using AI for.

A consumer using AI for discovery encounters more brands and may buy from new ones — Finding 1. The same consumer using AI for shopping assistance during a specific purchase may add more items as recommendations land — Finding 2. The same consumer using AI for research and comparison before a considered purchase may buy fewer items with higher confidence — Finding 3. Three use cases, three behaviors, same underlying technology.

Retailers reading AI behavior as a single amplification trend miss the structural shift. Demand planning calibrated to “AI makes consumers buy more” misses the deliberation pattern. Brand strategy assuming “AI commoditizes discovery” misses the conviction-building pattern. Customer experience optimizing for impulse misses the deliberate buyer. Last-mile architecture planning for volume growth misses the basket composition and returns pattern shifts.

Retailers reading AI behavior as three simultaneous patterns find the signal. Demand planning models incorporate both amplification and deliberation. Brand teams design for consideration breadth AND decision confidence. Customer experience architectures serve discovery, recommendation, and research use cases distinctly. Last-mile operations adjust to basket composition shifts, returns pattern evolution, and carrier mix complexity simultaneously.

The strategic question for enterprise leaders evaluating 2026 architecture is concrete: does the operational and commercial architecture read AI consumer behavior as three simultaneous patterns — wider consideration, larger baskets, and more confident deliberation — or as a single amplification trend that misses the structural shift in how US consumers now buy?

FAQs

What did Locus’s Q2 2026 US Consumer Survey find about AI shopping behavior?

Locus’s Q2 2026 US Consumer Survey finds AI users behave materially differently from non-AI users across three connected dimensions. AI users are more than 2x more likely to try new brands (39% vs 18%), more than 2x more likely to put more items in their carts (37% vs 17%), and more than 3x more likely to feel confident buying fewer items (34% vs 11%). The findings are not contradictory — they represent three behaviors happening simultaneously depending on what shoppers use AI for.

How does AI change consumer brand consideration?

AI-assisted shopping widens the consideration set materially. Consumers using AI tools encounter brands they wouldn’t have found through traditional search-and-browse patterns. 39% of AI users say they’re likely to try new brands compared with 18% of all consumers — more than twice the baseline rate. The shift affects brand defensibility for incumbents and creates discovery opportunity for challenger brands competing through AI-mediated channels.

Why do AI users put more items in their carts?

AI assistants function as recommenders during shopping flows, and the recommendations are landing. Bundled recommendations, complementary products, accessories, and alternatives surface during AI-assisted shopping in ways that lift basket size. 37% of AI users buy more items per order compared with 17% of all consumers. Discovery friction drops, recommendation relevance improves, and basket size rises as a result.

Why are some AI users buying fewer items?

Some consumers use AI for deliberation rather than amplification — comparing options more rigorously, researching alternatives, reaching higher confidence in fewer purchases. 34% of AI users feel confident buying fewer items compared with 11% of all consumers — more than 3x the baseline. The same AI shopping technology supports both amplification (Finding 2) and deliberation (Finding 3) depending on shopper use case.

What do the three findings mean for demand planning?

Demand planning models trained on pre-AI consumer behavior may underestimate the simultaneous patterns AI users exhibit. Brand consideration widens (affecting incumbent volume assumptions), basket size rises (affecting fulfillment economics per order), and deliberate purchasing patterns shift returns and frequency. Demand planning needs to handle amplification AND deliberation simultaneously rather than calibrating to a single trend direction.

How does AI consumer behavior affect last-mile operations?

Last-mile operations face several connected shifts. Expanded brand consideration produces wider carrier mix complexity as consumers buy from more shipping origins. Larger basket sizes change package mix, vehicle utilization patterns, and handling requirements. More deliberate purchasing typically produces lower returns volume but higher value per return. Last-mile architecture must adjust to basket composition shifts, returns pattern evolution, and carrier mix complexity simultaneously.

How should enterprise leaders read AI consumer behavior findings?

Enterprise leaders should read AI behavior as three simultaneous patterns rather than as a single trend. Wider consideration sets, larger baskets when buying, and more confident deliberation when not — three behaviors happening at once depending on shopper use case. Reading the data as single-trend amplification misses the deliberation pattern; reading it as three behaviors finds the structural signal that’s reshaping demand patterns, brand defensibility, and operational economics.