What really changes when a shopping AI stops picking products for you and starts haggling with the seller's own AI?
What makes agent-to-agent negotiation fundamentally different from agent-assisted shopping?
This explores what changes when an AI agent stops just helping a person shop (picking, comparing, buying for them) and starts negotiating with another AI agent that represents the other side of the deal.
This explores what changes when an AI agent stops just helping a person shop and starts bargaining with another agent that represents the seller. The short version from the corpus: shopping assistance is about steering one person's choice inside a market someone else built. Agent-to-agent negotiation changes the market itself: which deals are worth attempting, how long bargaining runs, and who ends up with the value.
In agent-assisted shopping, the main risk is that the agent is easier to sway than the person it serves. When the same price, rating, and nudge cues were shown to 17 models and to humans, the agents shifted their choices far more and fell into rule-like patterns, while people shifted only a little Do shopping agents respond more strongly to price and rating cues?. The influence also runs the other way. Agents that actually complete purchases don't just carry out what you already wanted. Repeated rounds of conversation help shape what you want Do AI shopping agents reshape what consumers actually want?. So the shopping problem is about where preferences come from and how easily they can be manipulated. The seller is mostly a storefront.
Negotiation puts an agent on both sides, and that changes the economics. Human bargaining is limited by time and attention. Agents are limited by compute, so they can run many negotiations at once, start earlier, and keep going much longer Why might AI agents bargain more patiently than humans do?. Choudary argues this opens up deals that never happened before because finding and haggling over them cost too much. Examples include Walmart's automated supplier negotiations and Honeywell's more than 2,500 autonomous deals Can agent negotiation unlock deals that seemed economically impossible?. Rothschild and colleagues at Microsoft take this further: the real effect may be lower switching costs and a shift in market power, not the automation of individual tasks Could agent-to-agent communication reshape market power?. A formal model adds a twist. A central recommendation algorithm has its biggest advantage in thin markets with few sellers, and that advantage shrinks toward zero as more sellers compete Does algorithmic recommendation lose its edge as sellers compete more?. If agents make thin markets thick, the advantage of a curated shopping assistant could fade.
The less obvious point is that in negotiation the quality of your agent decides your result, and a weak agent costs you money. In Anthropic's Project Deal, real people traded real goods for real money. Opus agents got about $2.68 more per item as sellers and paid about $2.45 less as buyers than Haiku agents did Do smarter AI agents negotiate better deals than weaker ones?. Being outgunned is also not the only risk. Across GPT, Qwen, and DeepSeek models, buyer agents revealed their budgets without being asked and overpaid, and seller agents accepted offers below wholesale. These were failures to follow instructions, not clever strategy Do LLM negotiation agents reliably follow budget and pricing rules?. A shopping agent that makes a mistake picks a worse product. A negotiating agent that makes a mistake gives away money to an opponent that may be better at the game.
Finally, negotiation needs infrastructure that shopping can mostly do without. Two agents dealing with each other have to prove who they are and what they're allowed to agree to. Amazon's block of Meta's Muse showed that this layer is still missing What identity standards do shopping agents actually need?. Emerging protocols like ACP, AP2, and Visa's TAP are building it up in layers Do agentic commerce protocols compete or complement each other?. Whether agents can talk freely or only within closed platforms will decide who captures the gains Will agentic commerce platforms adopt open or closed models?. Travel suggests the middlemen won't simply disappear. Online travel agencies are becoming the settlement layer that agents run on Are AI agents actually removing online travel agencies from bookings?.
Sources 12 notes
A benchmark testing 17 LLM models against humans on binary product choices found agents exhibited much larger choice shifts in response to price, rating, and nudge interventions—collapsing into rule-like patterns rather than the modest probabilistic shifts humans showed.
Bilgihan and colleagues argue that transactional AI agents—which execute purchases rather than just recommend—end the consumer's sole cognitive role in preference formation. Preferences emerge through bidirectional feedback loops between agent learning and consumer trust, not independently beforehand.
Shahidi and colleagues argue that unlike humans, AI agents face no time scarcity cost and can hold many parallel negotiations indefinitely. This compute-first economics means agents can negotiate earlier, concurrently, and longer than human bargainers, creating a structural advantage in multi-party negotiations.
Choudary argues bilateral agent representation makes possible deals that would never happen in human-run markets due to high discovery costs. Evidence includes Walmart's autonomous supplier negotiations, Honeywell's 2,500+ autonomous deals, and Anthropic's Project Deal completing 186 real agreements without human principals.
Microsoft researchers argue the deeper economic impact of agentic AI lies not in task automation but in reducing communication frictions between consumers and businesses. This shift could expand consumer options and reorganize markets, though the paper provides conceptual rather than empirical support for these claims.
Show all 12 sources
A formal model shows the welfare ratio of optimal algorithmic recommendations to Bertrand competition strictly decreases with seller count and approaches one as markets grow thick. The algorithm's edge exists only in thin markets where it can withhold trades to limit information rents.
Claude Opus agents extracted $2.68 more per item as sellers and paid $2.45 less as buyers than Haiku agents over identical marketplace transactions with real humans and real money.
Across GPT, Qwen, and DeepSeek models, buyer agents disclosed budgets unnecessarily, accepted inflated prices, and ignored cost limits, while sellers accepted below-wholesale offers. All four anomalies traced to instruction-following failures rather than strategic behavior.
Amazon's block of Meta's Muse reveals that AI agents need standardized ways to identify themselves and prove authorization to retailers—separate from consumer-side credential security. Without such protocols, retailers enforce access through terms-of-service blocks rather than negotiated standards.
Boboev argues ACP, AP2, UCP, TAP, and Web Bot Auth each solve different layers—application, governance, discovery, and identity—creating a capability ladder merchants stack rather than choose between, with each layer absorbing different levels of fraud risk.
Waller argues agentic commerce platforms will divide between open systems allowing multiple agents and closed systems requiring proprietary ones, mirroring payment network structure. This choice will shape market concentration, vertical integration, and standards adoption.
Goswami's 2026 travel data shows independent hotels' OTA dependence rose to 63.4% even as AI agents emerged, with OTAs retaining merchant-of-record status through API integrations. Expedia's B2B segment grew 24% year-over-year, now 38% of revenue.
Papers this line draws on 8
The research behind the notes this line reads — ranked by how closely each paper relates.
- When Agents Shop for You: Role Coherence in AI-Mediated Markets
- The Automated but Risky Game: Modeling and Benchmarking Agent-to-Agent Negotiations and Transactions in Consumer Markets
- From Product Search to Preference Articulation: The Economics of Agentic Commerce
- What Is Your AI Agent Buying? Evaluation, Biases, Model Dependence, & Emerging Implications for Agentic E-Commerce
- Agentic markets
- The Coasean Singularity? Demand, Supply, and Market Design with AI Agents
- Advancing AI Negotiations: A Large-Scale Autonomous Negotiation Competition
- LLM-Agent Interactions on Markets with Information Asymmetries