AI-DAPT is an EU-funded innovative research project focused on addressing critical challenges in AI deployment, particularly related to data utilization, model reliability, and adaptability. It aims at reinstating the pure data-related work in its rightful place in AI and at reinforcing the generalizability, reliability, trustworthiness and fairness of Al solutions.

• The project aims at reinstating the importance of data in AI, supporting automating data delivery for adaptable AI pipelines, while keeping the Human in the loop.  
• The project introduces smart and trustworthy adaptation in the Data/AI Ops lifecycle providing end-to-end automation and AI-driven methods to support the design, execution, and lifecycle management of data-AI pipelines. 
• The project harnesses sophisticated Explainable AI (XAI)-driven data operationsmethods to trigger for comprehensive data operations, ensuring transparency and accountability throughout the process. 
• AI-DAPT is driving AI innovation with hybrid science-guided and AI-based models, enhancing predictive accuracy and adaptability to diverse datasets, when scientific knowledge is present. 
• Demonstrating tangible innovation, AI-DAPT validates its results in real-world applications across industries like Health, Robotics, Energy, and Manufacturing, while integrating its advancements into existing AI solutions for market impact. 

Main challenges are the following:

• Challenge 1: Poor data preparation and planning/scoping, including delays in obtaining vital data or in requesting data too expensive to acquire or store or irrelevant 
• Challenge 2: “Messy“ data in terms of heterogeneous/ contradicting/ redundant values, missing entries, and inconsistent structure 
• Challenge 3: Data is not self-explanatory, well described and suitable for use beyond its original concept 
• Challenge 4: Low re-usability of data, features and models 
• Challenge 5: Representative data that are appropriate for AI may be hard to access or not be even available for sporadic events/edge cases​ 
• Challenge 6: Synthetic data needs to become more accurate​ 
• Challenge 7: Detection of bias and mitigation is a complex issue​ 
• Challenge 8: AI models developed “in vitro” are having a hard time in the real world 
• Challenge 9: Insights gained through data/AI observability are not yet actionable enough to trigger appropriate remedy actions​ 
• Challenge 10: XAI methods are interpretable only by data scientists​ 
• Challenge 11: Balanced use of Physics-based Models and AI is necessary to some problems 

AI-DAPT aims to deliver an innovative and impactful research agenda that will provide tangible benefits to a variety of stakeholders that struggle with making AI services. Seeking to reinstate the pure data-related work in its rightful place, and reinforcing the generalizability, reliability, trustworthiness, and fairness of Al solutions, AI-DAPT vision relies on the implementation of an AIOps framework to support and automate AI pipelines that continuously learn and adapt based on their context. It enables proper purposing, collection, documentation, (bias) valuation, annotation, curation and synthetic generation of data, while keeping humans-in-the-loop across five axis: (i) Data Design for AI, (ii) Data Nurturing for AI, (iii) Data Generation for AI, (iv) Model Delivery for AI, (v) Data-Model Optimization for AI.  

AI-DAPT brings forward a two-fold data-centric mentality in AI:  

• Data: AI-driven automation for data pipelines based on Explainable AI (XAI) techniques as well as synthetic data generation and observability.  
• Model: Automation on AI model building and hybrid science-AI solutions, bringing together data-driven AI models and science-based (first-principles) models that build on high-quality data.

Bridging the gap between data-centric and model-centric AI, AI-DAPT will turn over a new leaf in trustworthy AI and will nurture an ecosystem involving all AI and data value-chain stakeholders. The aim is to enhance their prosperous collaboration in order to deliver and apply innovative AI-driven methods that rely on smart and dynamic end-to-end automation of data, AI training/inference pipelines in the cloud-edge computing continuum. 

To demonstrate the actual innovation and added value that can be derived through the AI-DAPT scientific advancements, the AI-DAPT results will be validated in two ways:  

• By applying them to tackle real-world challenges in four key industries: (4) Health, Robotics, Energy, and Manufacturing. 
• By integrating them into various AI solutions, whether open source or commercial, already present in the market.