Camilla, let’s start with your background. What drew you to negotiation in the first place, and how did you end up exploring the role of AI in this area?
Camilla:
It’s a story that goes back quite a long way—and, in some ways, it started with a bit of contrast. Both of my parents work in sales, so I grew up surrounded by the day-to-day stories of negotiations, but always from the seller’s perspective. That naturally sparked my curiosity: I wanted to understand what was happening on the other side of the table—how decisions are made, and what kind of reasoning drives buyers.

That curiosity evolved during my academic journey in management engineering and eventually became a more structured research direction during my PhD. Over time, I began to observe how digital technologies—and AI in particular—were transforming decision-making processes. That’s when a new world opened up for me: negotiation, which is by nature a relational and dynamic activity, is undergoing deep changes driven by AI. And not just in operational terms, but in its epistemological and ethical foundations as well. That has become the core of my research.

Negotiation and AI seem like very different worlds. How can artificial intelligence play a role in something so inherently human?
That’s a fair point—they do seem quite distant at first glance. Negotiation is full of emotions, perceptions, intuition. But it’s also made up of sequences of decisions, trade-offs, and strategic preparation under uncertainty. This is exactly where AI can step in: collecting and analyzing data, generating scenarios, and suggesting strategies that buyers can use to optimize resource allocation.

It’s not about replacing the human negotiator—it’s about augmenting their decision-making capacity. In this sense, AI helps turn the buyer’s intelligence into a form of “augmented intelligence,” supporting preparation with market and counterpart analysis and helping identify the best strategies to achieve goals. In the near future, it could also provide real-time insights during negotiations, spot patterns or anomalies, or even autonomously handle certain negotiations through a chatbot. Of course, this all requires deep integration between human and algorithmic capabilities—and that’s one of the most fascinating areas of research right now.

You’ve mentioned that some large companies already use AI agents to negotiate with non-critical suppliers. How much automation is realistically possible in these kinds of scenarios?
Automation makes a lot of sense when negotiations are less complex or have low strategic value. We’re talking about non-critical suppliers, routine purchases where the negotiation parameters are highly standardized. In these cases, AI agents can handle negotiations autonomously, increasing efficiency and ensuring more consistent processes.

What’s interesting is that even suppliers—despite not being strategically critical—often report a better experience: they receive faster responses and enter a more transparent and consistent communication flow. This shows that automation doesn’t necessarily mean impersonal interaction, as long as it’s well designed.

Of course, as the strategic importance of a negotiation grows, human input becomes essential again—especially when advanced relational skills are required. But even in these cases, AI can still offer valuable support.

In these more complex scenarios, could an AI agent take part in the negotiation process in a different kind of role?
We’re not quite there yet in terms of widespread evidence, but in the near future—absolutely. AI could become a crucial decision-support tool at the negotiation table—a sort of context analyst. Just think about sentiment analysis: AI could detect subtle signals, changes in tone, recurring patterns in the counterpart’s responses. Or it could generate real-time scenario-based alternatives, giving the buyer a broader range of optimized options.

This kind of support—especially for strategic negotiations—isn’t about replacing the human but complementing them. It becomes an extension of the decision-maker’s cognitive capabilities, with the human still fully in control. This is where the idea of Hybrid Intelligence comes in, which I see as one of the most promising areas for the future. The key is enabling collaboration between human and artificial intelligence—allocating tasks according to their respective strengths and building a truly hybrid intelligence model.

But this vision also raises some ethical concerns. Is it really fair to introduce automated negotiation in contexts where there’s a power imbalance?
That’s a crucial question. Imagine a multinational company using AI to negotiate with small businesses: if the AI is solely optimized to maximize the buyer’s advantage, we risk exacerbating existing imbalances. Technological asymmetry can amplify economic asymmetry, potentially creating unethical or unsustainable dynamics over the long term.

That’s why we need clear ethical—and even regulatory—frameworks. Companies and policymakers alike need to ask: What are the boundaries of acceptable negotiation? What parameters should we put in place? How do we ensure that AI respects values like fairness, transparency, and sustainability? These are complex but absolutely necessary conversations.

Let’s go back to the buyer. In this increasingly tech-driven environment, how does their role evolve?
The buyer remains at the center—still the one holding the compass when it comes to using AI. They guide the collaboration with artificial intelligence, setting objectives, interpreting data, and using AI as a learning tool before the negotiation even starts. They also decide when to delegate autonomous negotiations to AI, identifying the right use cases and ensuring alignment between the company’s tech strategy and its procurement strategy.

AI can offer support, but the orchestration of the process still lies in human hands. And this balance—between human and artificial—is where the real challenge lies. A skilled buyer who understands the potential of AI and knows how to read the signals it provides can gain a significant competitive advantage. But they also need to be able to question the process, refine it, and make adjustments when needed.

In other words, it’s the buyer’s critical thinking that teaches the system. We’re nowhere near a point where that kind of critical evaluation can be left to a non-human intelligence. That’s still a cornerstone every manager must protect at all costs.

Given this shift, what kind of skills do you think buyers will need to develop over the next decade?
I see the buyer’s profile evolving in three main directions.

The first is technical: they’ll need a foundational understanding of AI models, the data they’re built on, and the limitations and biases they might contain. This knowledge is essential to using AI tools responsibly and ethically.

The second is relational and strategic: even as tech reduces human interaction, relational skills will remain absolutely critical. In an unstable geopolitical context, building and maintaining strong supplier relationships becomes essential. Buyers need to cultivate partnerships and steer negotiations toward collaboration rather than competition.

The third is educational: buyers must embrace lifelong learning. Not long ago, ChatGPT didn’t even exist and AI was largely the domain of developers—most users saw it as a “black box.” The rise of generative AI has radically changed that, making the tech much more accessible. But things are evolving fast, and only those who stay up to date will keep their seat at the table. That’s especially important because the counterpart in a negotiation will also have access to AI tools. Staying current can give you an edge—or at the very least, prevent you from falling behind.

To wrap up: what do you think are the most promising research directions in this field for the coming years?
I’d highlight two key areas. The first is ethics: we need a deeper understanding of how AI affects negotiations, particularly where there’s an imbalance in power or access to information. We need transparent governance models and new metrics to evaluate the fairness of deals.

The second is skills: we need to figure out which competencies will truly be strategic. That means rethinking training and career development programs through the lens of human–machine co-evolution. It’s time to reshape how we prepare the buyers of tomorrow.