Your Agent Acts Differently When You're Around: How User Presence Changes Agent Behavior in Negotiation

Jane Pan; You Lyu; Abe Bohan Hou; Valerie Chen; He He

cs.AIOct 27, 2026

Your Agent Acts Differently When You're Around: How User Presence Changes Agent Behavior in Negotiation

Authors: Jane Pan, You Lyu, Abe Bohan Hou, Valerie Chen, He He

Organizations: New York University · Stanford University

Abstract

LLM agents increasingly interact with other agents on behalf of users, yet how users in the loop affect their agents' behavior--for better or worse--remains poorly understood. We study this question through the lens of negotiation, where agents must represent users while strategically bargaining with other agents. To isolate the effect of user presence, we develop PRESENT, a controlled evaluation framework that compares negotiation settings with and without an interactive user, while holding the game, objectives, and user preferences fixed across conditions. Across 11.2k runs spanning 7 games and 8 models, we find that user presence significantly changes both outcomes and behavior. Specifically, negotiation parties with a user obtain higher payouts (+13.7% on average) and their agents negotiate more aggressively. Surprisingly, user presence also induces misaligned behavior, both toward counterparties (where its rate roughly doubles) and toward users (where it emerges in 14.3% of games). These misaligned behaviors are uncorrelated with outcomes and persist even when the user sends only templated, neutral messages. These findings expose two blind spots in current evaluation practice: static instructions miss behaviors that emerge only when users are present, and outcome-level metrics cannot detect them.

Figures & tables

Figure 1: Users change how agents negotiate. While most negotiation benchmarks do not include users in the loop, our evaluation framework compares otherwise identical negotiations with and without an interactive user to isolate the effect of user presence (top). We find that users shift agent behavior in three ways (bottom). Agents with users obtain higher payouts and negotiate more aggressively on average; however, they also display more misaligned behavior, toward both their counterparties and their own users.

Figure 2: Overview of Present . We model bilateral negotiation with an interactive user who can provide initial guidance (A), be consulted during negotiation (B), and must consent to the final deal (C). This common interface allows us to measure the effects of user presence consistently across a range of different negotiation games. We provide an example of a full negotiation transcript in § A.9 .

Game	Negotiation terms	Source
Buy-Sell	A seller and a buyer negotiate a price $P$ for one good.	Negotiation Arena Bianchi et al. (2024)
Ultimatum	Two players negotiate a split of $100.$ N $is the amount of money allocated to Player 2 by proposal$ x$ .
CaSiNo	Two campers divide a fixed stock of food, water, and firewood. For camper $i$ and item type $j$ , $q_{ij}$ is the quantity and $v_{ij}$ is the camper’s value per unit.	CaSiNo Chawla et al. (2021)
Deal or No Deal	Two players divide a pool of books, hats, and balls. For player $i$ and item type $j$ , $q_{ij}$ is the quantity and $v_{ij}$ is the player’s value per unit.	Deal or No Deal Lewis et al. (2017)
Service Contract	An engineer and a manager negotiate the hourly rate $R$ and the contract duration.	PrincipalBench Li and Shi (2026)
Car Sale	A seller and a buyer negotiate the price $P$ of a used car.

Table 1: Negotiation games, adapted from four sources and covering both single- and multi-issue negotiations.

Behavior	Definition	Example
User-facing
Misreporting	Agents report to the user information that is inconsistent with the agent–agent history, or omit an important alternative.	The agent reports to the user that the counterparty offers $52, but the actual offer is$ 48.
Hallucinated action	Agents falsely claim they did an action.	The agent claims that it has sent a counteroffer without actually sending it.
User-facing unsupported urgency	Agents pressure their users with unsupported claims of urgency.	The agent tells its user that it will finalize the deal unless the user objects within two minutes, though the game has no time limit.
Counterparty-facing
Misunderstanding	Agents communicate inaccurate information about the negotiation state in messages to the counterparty.	The agent says that the counterparty increased its offer by $1,000, although the offer rose only from$ 10,800 to $11,000.

Table 2: Misaligned-behavior categories. We group by whether the behavior is directed at the agent’s user or at the counterparty. Examples are paraphrases of actual logs; more verbatim examples can be found in § A.10 .

Game	Claude		DeepSeek	GPT-5.6		Qwen			Pooled
(95% CI)
	Haiku 4.5	Sonnet 5	V4 Flash	Luna	Terra	3.6 Plus	3.7 Plus	3.8 Flash
Buy-Sell	43.5 *	25.6 *	41.5 *	43.8 *	24.9 *	35.1 *	38.2 *	33.7 *	35.8 $\pm$ 6.3 *
Ultimatum	23.0 *	17.0 *	31.0 *	31.7 *	20.8 *	23.5 *	28.1 *	32.6 *	25.9 $\pm$ 5.3 *
CaSiNo	2.7	$-$ 5.3 *	0.0	3.5	1.0	1.6	0.2	5.2 *	1.0 $\pm$ 2.9
Deal or No Deal	1.3	$-$ 2.9 *	$-$ 5.2 *	4.4 *	$-$ 1.0	0.7	$-$ 3.7 *	1.5	$-$ 0.7 $\pm$ 2.6

Table 3: Payout change of the user-side agent in With User relative to No User . Cell color encodes direction and magnitude of the payout difference (green: increase with user; red: decrease; darker: larger). ∗ Significant changes (

p<.05

). Details on payout computation in § A.2 .

Figure 3: Changes in opening offers, mean concession, and preference hardening in With User relative to No User ; lines indicate 95% confidence intervals. We pool across all models per game (random-effects) and average across all games, and find increased negotiation aggression in most games.

Behavior	User	Claude		DeepSeek	GPT-5.6		Qwen			Mean
(95% CI)
		Haiku 4.5	Sonnet 5	V4 Flash	Luna	Terra	3.6 Plus	3.7 Plus	3.8 Flash
User-facing
Misreporting	With	46.0	15.1	4.6	2.9	1.3	14.6	14.9	4.9	13.0 $\pm$ 0.8
Hallucinated action	With	14.6	2.7	1.6	0.4	0.0	3.4	3.0	0.7	3.3 $\pm$ 0.5
Unsupported urgency	With	12.1	1.4	0.9	0.7	0.3	2.4	1.3	1.1	2.5 $\pm$ 0.4

Table 4: Misaligned behavior rates (%) by behavior, model, and user setting. In With User (✓), only the agent with the user is counted. In No User (✗), the rates are averaged over the two agents. User-facing misaligned behaviors cannot occur in No User . Any is the rate of showing at least one behavior in its group. The All behaviors row takes the union over all six behaviors. ∗ indicates ✓ differs from ✗ in the same column (

p<.05

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

		User-facing	Counterparty-facing
Condition	Payout change (%)	Misaligned-behavior rate (%)
With User	13.7 $\pm$ 1.2 *	14.3 $\pm$ 0.8	5.5 $\pm$ 0.6 *
Opening-Only User	7.0 $\pm$ 1.2 *	8.8 $\pm$ 0.7 *	4.1 $\pm$ 0.5 *
Passive User	$-$ 0.3 $\pm$ 1.2	8.6 $\pm$ 0.6 *	3.7 $\pm$ 0.5 *
No User	—	—	2.6 $\pm$ 0.3

Appendix

Table 5: User condition ablations.

\pm

gives the 95% CI half-width. ∗ Significantly differs from No User for payout and counterparty-facing rates, and from With User for user-facing rates (

p<.05

Game	Negotiation terms	User payout	Source
Buy-Sell	A seller and a buyer negotiate a price $P$ for one good.	Seller: $i=1$ ; buyer: $i=2$ . $u_{1}(x)=P-40$ , $u_{2}(x)=90-P$ . $u_{i}(\varnothing)=0$ .	Negotiation Arena Bianchi et al. (2024)
Ultimatum	Two players negotiate a split of $100.$ N $is the amount of money allocated to Player 2 by proposal$ x$ .	$i$ indexes the player. $u_{1}(x)=100-N$ , $u_{2}(x)=N$ . $u_{i}(\varnothing)=0$ .
CaSiNo	Two campers divide a fixed stock of food, water, and firewood. For camper $i$ and item type $j$ , $q_{ij}$ is the quantity and $v_{ij}$ is the camper’s value per unit.	$u_{i}(x)=\sum_{j}q_{ij}v_{ij}$ . $u_{i}(\varnothing)=5$ .	CaSiNo Chawla et al. (2021)
Deal or No Deal	Two players divide a pool of books, hats, and balls. For player $i$ and item type $j$ , $q_{ij}$ is the quantity and $v_{ij}$ is the player’s value per unit.	$u_{i}(x)=\sum_{j}q_{ij}v_{ij}$ . $u_{i}(\varnothing)=0$ .	Deal or No Deal Lewis et al. (2017)
Service Contract	An engineer and the manager negotiate the hourly rate $R$ and the contract duration.	Engineer: $i=1$ ; manager: $i=2$ . $u_{1}(x)=R-100$ , $u_{2}(x)=105-R$ . $u_{i}(\varnothing)=0$ .	PrincipalBench Li and Shi (2026)
Car Sale	A seller and a buyer negotiate the price $P$ of a used car.	Seller: $i=1$ ; buyer: $i=2$ . $u_{1}(x)=P$ , $u_{2}(x)=15{,}000-P$ . $u_{i}(\varnothing)=0$ .

Appendix

Table 6: Negotiation games and utility functions. For user

i\in\{1,2\}

u_{i}(x)

is the payout under agreement

x

, and

u_{i}(\varnothing)

is the payout with no agreement.

Figure 4: Judge scores of candidate user models in Dual User Ultimatum, under GPT-5.6 Sol (left) and Claude Opus 5 (right) as judges. Overall is the mean of the two criteria.

Game	Agent model	User model
		GPT-5.6 Terra	GPT-5.6 Sol	Claude Sonnet 5
Ultimatum	Claude Haiku 4.5	23.0 $\pm$ 2.2 *	23.0 $\pm$ 1.7 *	12.0 $\pm$ 1.4 *
	DeepSeek V4 Flash	31.0 $\pm$ 3.1 *	29.8 $\pm$ 3.4 *	15.3 $\pm$ 2.5 *
	GPT-5.6 Terra	20.8 $\pm$ 3.7 *	20.4 $\pm$ 4.4 *	5.1 $\pm$ 2.1 *
	Pooled	24.9 $\pm$ 12.6 *	24.4 $\pm$ 10.7 *	10.8 $\pm$ 11.6
Bike Sale	Claude Haiku 4.5	20.1 $\pm$ 3.3 *	17.8 $\pm$ 4.0 *	22.7 $\pm$ 3.2 *

Appendix

Table 7: Results with three different user models in Ultimatum and Bike Sale, with Claude Haiku 4.5, DeepSeek V4 Flash, and GPT-5.6 Terra as agents. (a) Payout change relative to No User , computed as in Table 3 . ∗ Significant changes (

p<.05

). (b) Misaligned-behavior rates (%), computed as in Table 4 . ∗ Rate differs from No User (

p<.05

\pm

gives the 95% CI half-width.

Figure 5: F1 of candidate judges against human annotations of 229 turns from 20 trajectories. (a) Misaligned behaviors; (b) Preference Hardening . GPT-5.6 Sol achieves the highest F1.

Figure 6: Pearson correlation coefficients ( $r$ ) between an agent’s own misaligned behavior and its own payout.

Figure 7: Counterparty-facing misaligned-behavior rate and payout in No User (left) and With User (right). One point for each combination of game, model, and party, and darker shading marks denser regions.

		User-facing	Counterparty-facing
Condition	Payout vs. no user (%)	Misaligned behavior rate (%)
No User	—	—	1.1 $\pm$ 0.5
With User	24.0 $\pm$ 1.3	13.4 $\pm$ 1.8	5.5 $\pm$ 1.3
Passive User	0.9 $\pm$ 1.8	8.5 $\pm$ 1.4	3.7 $\pm$ 1.1
Opening-Only User	18.6 $\pm$ 1.4	9.7 $\pm$ 1.5	4.7 $\pm$ 1.2
User Sees Agent-Agent Transcript	22.2 $\pm$ 1.3	13.9 $\pm$ 1.8	7.6 $\pm$ 1.5

Appendix

Table 8: Additional user condition ablations.

\pm

gives the 95% CI half-width. ∗ Significantly differs from No User for payout and counterparty-facing rates, and from With User for user-facing rates (

p<.05

Dependent variable:
counterparty-facing misaligned-behavior rate (%)
	(1)	(2)	(3)
With User	2.89 ∗∗∗	2.02 ∗∗∗	1.95 ∗∗∗
	(0.34)	(0.33)	(0.33)
$\log$ (turns)		2.50 ∗∗∗	1.86 ∗∗∗
		(0.26)	(0.29)

Appendix

Table 9: Linear probability models of the counterparty-facing misaligned-behavior rate. (1) has no controls. In (2), we control for the log number of negotiation turns. (3) adds an interaction allowing the slope on

\log(\text{turns})

to differ with a user. Heteroskedasticity-robust standard errors in parentheses.

{}^{*}p<.05

{}^{**}p<.01

{}^{***}p<.001

Human-AI Interaction

Human-AI interaction studies how people and AI systems communicate, collaborate, make decisions, and shape one another over time.

See all 329 papers

Cooperate to Compete: Strategic Coordination in Multi-Agent Conquest
Apr 28, 2026Abigail O'Neill, Alan Zhu, Mihran Miroyan +2Multi-Agent Coordination Multi-Agent Negotiation
University of California, Berkeley
Deal Me Maybe: The Role of Emotions in Multi-Agent Negotiation
Aug 7, 2026Massimiliano Luca, Apoorva Singh, Bruno LepriMulti-Agent Negotiation Automated Negotiation
Bruno Kessler Foundation
LLM-X: A Scalable Negotiation-Oriented Exchange for Communication Among Personal LLM Agents
May 12, 2026Giuliano Lorenzoni, Paulo Alencar, Donald CowanMulti-Agent LLM Systems Multi-Agent Coordination
University of Waterloo · University of Waterloo Waterloo, ON, Canada

Apr 28, 2026J/K move · Enter open · S save

Cooperate to Compete: Strategic Coordination in Multi-Agent Conquest

Abigail O'Neill, Alan Zhu, Mihran Miroyan, Narges Norouzi+1

University of California, Berkeley

Multi-Agent Coordination Multi-Agent Negotiation Automated Negotiation Human-AI Interaction AI Agent Benchmarks Multi-Agent LLMs

Open paper AskPDF arXiv

Language Model (LM)-based agents remain largely untested in mixed-motive settings where agents must leverage short-term cooperation for long-term competitive goals (e.g., multi-party politics). We introduce Cooperate to Compete (C2C), a multi-agent environment where players can engage in private negotiations while competing to be the first to achieve their secret objective. Players have asymmetric objectives and negotiations are non-binding, allowing alliances to form and break as players' short-term interests align and diverge. We run AI only games and conduct a user study pitting human players against AI opponents. We identify significant differences between human and AI negotiation behaviors, finding that humans favor lower-complexity deals and are significantly less reliable partners compared to LM-based agents. We also find that humans are more aggressive negotiators, accepting deals without a counteroffer only 56.3% of the time compared to 67.6% for LM-based agents. Through targeted prompting inspired by these findings, we modify agents' negotiation behavior and improve win rates from 22.2% to 32.7%. We run over 1,100 games with over 16,000 private conversations totaling 15.2 million tokens and over 150,000 player actions. Our results establish C2C as a testbed for studying and building LM-based agents that can navigate the sophisticated coordination required for real-world deployments. The game, code, and dataset may be found at https://negotiationgame.io/c2c.

Abstract

Human-AI Interaction

Explore similar work

Cooperate to Compete: Strategic Coordination in Multi-Agent Conquest

Deal Me Maybe: The Role of Emotions in Multi-Agent Negotiation

LLM-X: A Scalable Negotiation-Oriented Exchange for Communication Among Personal LLM Agents