Imagine a GPS that doesn't just tell you where to turn, but explains why and listens when you disagree. Most AI acts like a black box, giving orders without explanation, which makes people ignore them. This work builds a two-way street where the AI learns from the person and the person trusts the AI because they actually understand its logic.
Companies struggle to get employees to trust and use AI for complex tasks because the AI is a 'black box' and doesn't account for human preferences. This leads to low adoption rates and wasted investment in automation.
A two-way communication system for human-AI collaboration and a quantitative AI acceptance index to measure trust and fairness.
If you are a fleet management provider dealing with drivers who ignore AI route suggestions — this project developed a two-way communication system that increases trust. It allows drivers to provide feedback and understand the AI's reasoning, leading to better adoption of optimized routes.
If you are a smart factory operator dealing with complex sequences of machine operations — this project developed a collaborative AI assistant. It ensures human operators and AI work together to control processes, reducing errors caused by a lack of transparency in AI decisions.
If you are an AI tooling vendor dealing with low user retention due to confusing interfaces — this project developed an AI acceptance index. This tool allows you to measure and improve how users perceive fairness and trust in your software.
Based on available project data, no specific pricing or commercial cost for the end product is mentioned; the EU provided a contribution of EUR 7,737,900 for the research phase.
The project is designed for industrial scale, as it integrates and evaluates its methods in 4 real-world use cases.
Based on available project data, specific licensing terms are not provided, though the project involves 9 industry partners who likely share in the results.
The project focuses on creating bidirectional conversation and collaboration tools that can be integrated into sequential decision-making systems like routing or process control.
The project period runs from 2023-10-01 to 2027-09-30, suggesting that final validated results will be available by late 2027.
The consortium is heavily weighted toward commercial application, with a 53% industry ratio consisting of 9 companies, including 4 SMEs. This strong industrial presence, combined with 4 universities and 3 research centers across 9 countries, suggests the results are being developed with direct market needs in mind rather than purely academic interest.
Contact Vrije Universiteit Brussel
Contact us to connect with the PEER consortium for early access to the AI acceptance index.
