🤖 AI-augmented negotiation
AI & decision support
AI-assisted negotiation: research on negotiating agents (N. Brown et al., CICERO, Science 2022), decision support and bias detection; preparation and debriefing with large language models.
Full detail in the “Origin & history” section below.
Author of the library
Negotiation lecturer at KEDGE Business School
A rare threefold background serving negotiation: teaching, commercial justice and audit, backed by experience as a company director in Bordeaux.
The Real-Time Negotiation Copilot is an artificial-intelligence assistant that listens to or reads the exchange as it unfolds (video call, phone, chat, synchronous email) and discreetly feeds suggestions to the negotiator: rewordings, responses to an objection, a reminder of the BATNA, detection of verbal and emotional signals. It turns conversation intelligence, until now post-mortem, into "in-the-moment" support that increases the negotiator's available cognitive bandwidth. Its value rests on the ability to process language faster than a human and to draw on a prepared knowledge base; its limit lies in the risk of dependence, of relational disconnection and of confidential data leaks. Well used, it is a "second brain"; badly used, an autopilot that makes you lose the room.
At a glance
Vigilance: moderate (4.0/10) · Preparation required: 8/10
Indicative profile: it situates the “Négociation augmentée par l'IA” family as the AI-augmented negotiation school practises it, not this technique taken in isolation. Techniques from the same family and school therefore share the same profile. NEGOCOACH editorial rating out of 10, non-experimental · the higher the “relational risk” value, the more costly the technique is to the relationship.
NEGOCOACH assessment
Tactical potential 5.7/10 (effectiveness, impact, discretion) and vigilance moderate (relational and ethical risk): two distinct readings, deliberately never merged into a single score that would reward risk. NEGOCOACH editorial rating calibrated from the “Négociation augmentée par l'IA” family and the “AI-augmented negotiation” school. Each criterion is rated out of 10; click to understand what it measures.
How far the technique can carry the negotiation in the intended direction when it is well executed.
Strength of the effect produced on the counterpart's perceptions, emotions and decisions.
How hard it is for the other party to notice the technique is being used. A high value = very discreet.
The information, analysis and rehearsal required upfront to use it effectively.
Potential cost to the relationship and to trust if the technique is spotted, refused or fails. A high value = riskier.
Moral acceptability: fairness, transparency and respect for the counterpart's autonomy. A high value = more defensible.
Documented school
The school this technique stems from is documented by recognised work and established practice, without experimental consensus. This indicator qualifies the school, not this technique taken in isolation.
Indicative NEGOCOACH editorial rating, for teaching purposes. For “Relational risk”, a high value signals a cost to the relationship, not a quality.
The technique has no single academic author: it is the transposition to negotiation of the real-time "conversation intelligence" born in contact centres and sales. The conceptual building blocks come from the virtual negotiating agents studied by Jonathan Gratch and David DeVault (USC Institute for Creative Technologies, 2010s) and from the work on automated negotiating agents by Tim Baarslag and Catholijn Jonker. The "live assist" industrialisation has been driven since ~2017-2020 by vendors such as Cresta (Agent Assist), Gong and Microsoft 365 Copilot, then, with generative AI (2023+), by consumer copilots specifically dedicated to negotiation (e.g. "negotiation copilot" demonstrators). The strongest scientific basis for the "live AI assistance" effect is the study by Brynjolfsson, Li and Raymond (2023) showing a productivity gain for agents assisted in real time.
A software system that, during an ongoing negotiation, captures the flow of the conversation (live audio transcription or text), analyses it against a prepared context (objectives, ZOPA, BATNA, the other side's profile, arguments) and delivers to the negotiator, over a private channel not visible to the other party (screen overlay, earpiece, side panel), actionable micro-recommendations: what to say now, how to reword, which concession to propose, which signal (hesitation, pressure tactic, emotion) has just appeared. Operationally, it combines speech recognition, natural-language processing, intention/emotion detection and response generation, under the control and arbitration of the human negotiator, who remains the sole decision-maker.
Application by context
During a closing video call, the copilot detects the price objection, instantly displays the costed value/ROI and suggests a trade-off (a multi-year commitment) rather than a flat discount, sparing the salesperson from caving under pressure.
Faced with a supplier invoking a rise in raw materials, the copilot recalls in real time the actual market index and proposes a bounded indexation formula, defusing the argument from authority with data.
During a tense annual pay round (NAO), the assistant flags to the HR director a calming reformulation and prompts them to acknowledge the union's emotion before responding on the figures, reducing escalation.
In a crisis management negotiation (product recall, a contract held commercially "hostage"), the copilot monitors the tone, warns of rising aggressiveness and proposes calibrated de-escalation and time-gaining phrases.
In a coalition negotiation, an adviser uses a private panel that tracks the positions already conceded by each party and warns when a proposal crosses a publicly announced red line, avoiding inconsistency.
On the phone with a seller, the assistant reformulates the prospective buyer's offer, recalls the comparable price per square metre in the neighbourhood and suggests the next minimal concession to give up to stay within the ZOPA.
In a French-Japanese exchange, the copilot signals that the counterpart's silence is not a refusal but a norm of reflection, and advises against filling the void with a premature concession.
During an inheritance mediation by video call, a relative discreetly relies on an assistant that helps them separate facts from emotions and reformulate everyone's needs without judgment, keeping the exchange constructive.
Counter-techniques
Negotiation is also played on defence. Here is how to recognise this technique when it is used against you, and turn it around.
The signals that give it away
The counters that defuse it
Turn it into an advantage
Name the manoeuvre: said out loud, a technique loses most of its power.
Reacting emotionally instead of coming back to the facts.
Strengths: lightens the mental load and secures preparation by making it available "at the right moment"; homogenises quality (arguments, rewordings) regardless of stress or fatigue; accelerates juniors' training through in-situ learning; captures signals (pace, vocabulary, emotions, tactics) faster than human attention; keeps an analysable trace for the debrief. Weaknesses: a risk of overload and latency (reading the screen = no longer listening to the other); generic or out-of-context suggestions that ring false and break authenticity; dependence that atrophies the negotiator's judgment; a blind spot on fine non-verbal cues and tacit context; confidentiality issues (the conversation leaves the room) and ethics if the other party is unaware of it; quality entirely conditioned by that of the preparation injected.
Most useful in negotiations with high technical or informational content (complex purchasing, SaaS contracts, tenders) where recalling data in real time makes the difference; remote exchanges (video call, phone, chat) where a discreet overlay is possible; repeatable sales closings where consistency of responses counts; the upskilling of junior negotiators; situations of high cognitive or emotional load where a safeguard (BATNA, red lines) protects against drift. Less suited to intimate face-to-face meetings, to highly relational/political negotiations where presence and improvisation take precedence, and when confidentiality forbids any recording.
Sales · The agent assist that boosts productivity in real time, In a large field study conducted at an enterprise-software vendor, more than 5,000 customer-support agents were equipped with a generative-AI assistant suggesting responses in real time during conversations. The authors (Brynjolfsson, Li and Raymond) measured a productivity increase of about 14% on average, concentrated among the least experienced agents, who reached the level of seniors more quickly. This case, although on the customer-service side and not negotiation strictly speaking, is the most solid empirical demonstration of the real-time copilot mechanism: diffusing the know-how of the best towards the less seasoned, in real time.
Business · The procurement copilot that reframes the supplier's argument (representative scenario), Representative scenario (not attributed to an identified organisation). A buyer at an industrial group negotiates by video call the renewal of a components contract. The supplier opens with a request for a 12% increase "because of the market". A copilot connected to the buyer's screen instantly displays the actual movement of the material index (+4%), flags the gap and prompts her with two options: ask for the price breakdown and propose a transparent indexation clause. The buyer no longer submits to the argument from authority, brings the discussion back to the facts and closes the gap. The case illustrates the typical contribution, data at the right moment, framing rather than a canned reply, without claiming a verifiable source.
Spotting its use by the other party: a gaze that systematically drifts towards a second screen, micro-latencies before every response, wordings that are suddenly very "clean", polished and impersonal, responses perfectly calibrated but disconnected from the emotion of the moment. To defend yourself: deliberately slow the pace (the AI helps most with speed), ask open, unexpected and personal questions that fall outside the prepared script, change channel (switch to in-person, cut the video call), and frontally ask for a spontaneous reaction. On the rules front, you can demand transparency about any recording/transcription. Finally, equipping yourself with a copilot restores informational symmetry.
Technical limits: speech recognition and emotion analysis remain fallible (accents, overlaps, irony), and generative AI can hallucinate a figure or a false argument at the worst moment; fine non-verbal cues and tacit context largely escape the system. Cognitive limits: an overload effect and dependence that erode autonomous judgment. Major ethical and legal stakes: live transcription processes personal and often confidential data (GDPR/RGPD, trade secrets), capture without a legal basis or notice to participants is problematic; the asymmetry created by hidden use raises a question of fairness in the negotiation; the conversation "leaving the room" towards a third-party service creates a leak risk. Guiding principle: the AI assists, the human decides and bears the responsibility; favour controlled solutions on sensitive matters and remain transparent when the context requires it.
Related techniques: AI-augmented debriefing (post-negotiation analysis, #AI family); the AI negotiation simulator (training before the exchange); AI-assisted preparation (generating BATNA/ZOPA, mapping interests); the autonomous negotiating agent that negotiates in your place (Baarslag/Jonker's automated negotiating agents); classic conversation intelligence (Gong) centred on after-the-fact analysis; live drafting assistance for written negotiations (email, chat). The real-time copilot sits between mere post-mortem analysis and full automation: the human keeps control, the AI augments.
Quick exercise
Answer in your head, then reveal the solution. Memory is built through active recall.
Frequently asked questions
The Real-Time Negotiation Copilot is an artificial-intelligence assistant that listens to or reads the exchange as it unfolds (video call, phone, chat, synchronous email) and discreetly feeds suggestions to the negotiator: rewordings, responses to an objection, a reminder of the BATNA, detection of verbal and emotional signals. It turns conversation intelligence, until now post-mortem, into "in-the-moment" support that increases the negotiator's available cognitive bandwidth. Its value rests on the ability to process language faster than a human and to draw on a prepared knowledge base; its limit lies in the risk of dependence, of relational disconnection and of confidential data leaks. Well used, it is a "second brain"; badly used, an autopilot that makes you lose the room.
It sits on the line: effective, but it can tip into manipulation if it exploits an information asymmetry. Use it with measure and without deliberate deceit.
Reacting emotionally instead of coming back to the facts. The right reflex: slow down and reformulate.
NEGOCOACH does not assess the experimental validation of this technique in isolation. What we document is the grounding of its source school (AI-augmented negotiation): documented school. Full detail is in the "At a glance" section of this page.
Practise with AI
Copy, paste into your assistant, replace the [brackets]. Works with ChatGPT, Claude, Gemini, Mistral, Perplexity.
Build your plan before the meeting
You are an expert negotiation coach. Help me prepare to use the "The Real-Time Negotiation Copilot" technique in the following situation: [describe your situation]. Give me: the conditions for success, a 3-step script, my counterpart's likely objections and how to answer them.
Rehearse against an AI counterpart
Play the role of my counterpart in a negotiation. I am going to test the "The Real-Time Negotiation Copilot" technique. React realistically and with resistance, do not give in too quickly, then at the end analyse my performance and suggest 3 concrete improvements.
Analyse a past negotiation
Here is how my negotiation went: [paste the exchanges]. Analyse whether the "The Real-Time Negotiation Copilot" technique was used well, what worked, the mistakes made, and spell out precisely what I could have done better.
References
Founding works of the 🤖 AI-augmented negotiation school this technique belongs to.
Human-level play in the game of Diplomacy (CICERO), Science
ArticleFAIR (Meta) - N. Brown et al. · 2022
Noise: A Flaw in Human Judgment
BookD. Kahneman, O. Sibony & C. Sunstein · 2021
Co-Intelligence: Living and Working with AI
BookE. Mollick · 2024
AI-assisted negotiation: research on negotiating agents (N. Brown et al., CICERO, Science 2022), decision support and bias detection; preparation and debriefing with large language models.
On video
Videos to picture The Real-Time Negotiation Copilot and anchor it through examples.
A verified video selection is being enriched; the search above already surfaces the best videos on the topic.
Technique map
Every technique sits within a network: what it draws on, what it combines with, where it applies, and how to defend against it.
Spot its signals, neutralise it and turn it around with the defensive playbook on this page.
See the counter-techniquesThe Real-Time Negotiation Copilot is an artificial-intelligence assistant that listens to or reads the exchange as it unfolds (video call, phone, chat, synchronous email) and discreetly feeds suggestions to the negotiator: rewordings, responses to an objection, a reminder of the BATNA, detection of verbal and emotional signals. It turns conversation intelligence, until now post-mortem, into "in-the-moment" support that increases the negotiator's available cognitive bandwidth. Its value rests on the ability to process language faster than a human and to draw on a prepared knowledge base; its limit lies in the risk of dependence, of relational disconnection and of confidential data leaks. Well used, it is a "second brain"; badly used, an autopilot that makes you lose the room.
Name the manoeuvre: said out loud, a technique loses most of its power.
Reacting emotionally instead of coming back to the facts.
Our programmes turn theory into a concrete advantage.