Kontext
Foodborne illnesses remain a major public health challenge in Europe, causing approximately 23 million cases of illness and 5,000 deaths each year. The EU’s Rapid Alert System for Food and Feed (RASFF) issues over 3,000 food safety alerts annually, highlighting recurring contamination in key food products like nuts, fruits, vegetables, and poultry. These risks disproportionately affect vulnerable groups—infants, the pregnant, the elderly, and the immunocompromised—exacerbating health inequalities.
Beyond the health burden, unsafe food disrupts trade, reduces consumer trust, and results in significant economic and environmental costs. As food supply chains become more complex, and climate change introduces leads to emerging food safety hazards, there is an urgent need for a more systemic, predictive approach to managing food risks.
HOLiFOOD responds to this need by initiating a holistic, integrated framework for food safety that moves beyond the traditional reactive approach. The project combines next-generation data analytics, AI-powered early warning systems, and co-created risk assessment tools to enhance foresight and rapid response capabilities across the supply chain. It promotes cross-sector collaboration between scientists, policymakers, producers, and consumers, strengthening Europe’s ability to anticipate and mitigate emerging threats.
Objectives
The overall objective of HOLiFOOD is to improve the food safety risk analysis framework in Europe to:
- Meet future challenges arising from Green Deal policy driven transitions, in particular in relation to climate driven changes.
- Support the realisation of a truly safe and sustainable food production.
- Contribute to the United Nations’ Sustainable Development Goals (SDGs 2, 8, 9, 12, 15).
Specifically the project aims to develop:
- Early Warning and Emerging Risk Prediction Systems to identify and monitor existing and emerging food safety risks (ERI) in the food chain.
- Targeted and non-targeted detection methods for existing and emerging hazards.
- Holistic risk assessment methods and tools to support regulation in a changing global environment.
- Improved data and knowledge sharing infrastructures by developing an Integrated European Data and Knowledge Exchange Infrastructure that will be able to power an ecosystem of decision support systems.
Activities
HOLiFOOD addresses the increasing complexity of food supply chains, climate-driven risks, and shifting consumer practices through advanced prediction, detection, and policy-aligned innovation.
- At the core of HOLiFOOD is a technology-driven early warning system, powered by AI and big data analytics. These tools automatically extract relevant data from public sources—ranging from climate and economic indicators to societal trends—and feed them into predictive models that anticipate emerging food safety risks. This allows the shift from reactive to proactive food safety management based on earlier interventions and better preparedness.
- HOLiFOOD is also advancing next-generation detection technologies. By developing and validating both targeted and non-targeted methods for identifying chemical and biological hazards, the project strengthens the ability to detect (re-)emerging risks across the food supply chain. These tools support rapid, reliable characterization of threats, helping to mitigate potential impacts before they escalate.
- Recognizing the need for system-wide insights, HOLiFOOD employs a holistic risk assessment framework. It evaluates food safety risks across different timescales (short to long-term) and from multiple dimensions, including risk-benefit, environment and economic, while taking into account climate change impacts. This work is grounded in a co-creation process involving diverse actors across three selected food chains, ensuring real-world relevance and uptake.
- To enhance knowledge flow and decision-making, HOLiFOOD is building an integrated digital infrastructure for food safety data and tools. This platform will harmonize inputs from public and private stakeholders, offering a shared registry of data sources, models, and risk indicators. It will act as a hub for proactive risk identification and systemic food safety management.
- Through a Living Lab approach, the project engages policymakers, producers, industry experts, and citizens to co-design and pilot food safety solutions. These labs operate both vertically (focused topics) and horizontally (cross-sector workshops), fostering shared learning and practical tool development.
- The project also emphasizes science-to-society and policy integration. HOLiFOOD translates scientific insights into actionable policy recommendations, regulatory tools, and public engagement strategies. By using citizen science methods—such as surveys, focus groups, and social media monitoring—it informs risk communication and ensures that messages resonate with diverse audiences across four countries.
Through its integrated activities, HOLiFOOD provides the knowledge, tools, and stakeholder engagement necessary to support safe, sustainable, and resilient food systems in Europe.
Project details
- Main funding source
- Horizon Europe (EU Research and Innovation Programme)
- Type of Horizon project
- Multi-actor project
- Project acronym
- HOLiFOOD
- CORDIS Fact sheet
- Project contribution to CAP specific objectives
-
- SO2. Increasing competitiveness: the role of productivity
- SO4. Agriculture and climate mitigation
- SO5. Efficient soil management
- SO9. Health, Food & Antimicrobial Resistance
- Environmental care
- Protecting food and health quality
- Fostering knowledge and innovation
- Project contribution to EU Strategies
- Reducing the overall use and risk of chemical pesticides and/or use of more hazardous pesticides
EUR 6 056 434.25
Total budget
Total contributions including EU funding.
EUR 6 056 434.25
EU contribution
Any type of EU funding.
Ressourcen
Audiovisual materials
8 Practice Abstracts
Through AI, an end-user can then assess the risks of specific food safety hazards in the product that they are interested in. One of the issues with these next-generation AI algorithms is their complexity, therefore considered to be a ‘black box’.
In the HOLIFOOD project, we developed multiple artificial intelligence (AI) methods to predict food safety hazards and emerging food safety risks in three supply chains.
We used a so-called holistic approach, which incorporates many different characteristics that might influence food safety hazards. These characteristics can relate for example to weather and climate change. To see what is going on under the hood of this AI algorithm, we employ explainable AI. This makes the AI explain why it came up with a decision.
With these methods we aim to:
- Enhance risk prediction capabilities through AI algorithms that consider multiple factors
- Reduce food safety incidents by predicting hazards before they occur.
The entire approach and methodology will be available here: https://holifoodproject.eu/
The growing threat of (re-)emerging foodborne hazards driven by factors like environmental change, global trade, and resource shortages pose increasing risks to food safety. Existing detection methods were often too slow, targeted, or limited to effectively predict, detect, and manage these risks across complex food chains.
Multiple innovative methods for detecting biological and chemical food hazards have been developed, including quasi-metagenomic approaches, WGS-based assays, biosensors, and AI-driven analysis techniques.
Throughout the project we combine targeted and untargeted detection methods, advanced molecular biology techniques, and AI-driven models, integrating results into a European data-sharing platform to support decision-making.
This technology offers:
- Reduced foodborne illness cases with advanced detection methods for pathogens and chemicals
- On-site, real-time detection capabilities with portable microfluidic devices
- Better understanding of how pathogens persist in food processing environments
The outcome of our studies will be available in: https://holifoodproject.eu/
For the future, we hope to see this research move forward and:
- Apply the knowledge about pathogen persistence to redesign processing environments that minimize contamination
- Establish testing protocols that combine both targeted and untargeted methods for comprehensive hazard detection.
HOLiFOOD implements a coordinated Living Labs (LL) approach, aiming to address the complexity of emerging food safety risks by fostering dialogue between science and society, ensuring solutions are both scientifically sound and socially accepted.
The 3 LLs focus on:
- AI-driven risk identification and monitoring
- Holistic risk assessment and acceptance
- AI-driven digital platform co-design
The LL methodology supports multi-actor engagement through a structured framework for interaction, testing, and validation of new tools in real-life contexts.
Drawing on a quasi-experimental approach, LLs follows three phases: exploring the current state and envisioning alternatives; experimenting with innovations; evaluating results to inform iterative development.
The key findings that emerge:
- LL Managers and Facilitators are critical to ensure inclusive, participatory processes and foster dialogue.
- Stakeholder engagement works best as an iterative process with regular feedback loops and informal meeting to realign goals.
- Engagement methods must be adaptive and supported by clear
- Centralised coordination across LLs enhance methodological consistency and enables shared use of tools and strategies.
Practical implications for practitioners for replicating the LL model include alternating online and in-person meetings to improve flexibility and participation; developing visual materials (infographics, info sheets) to support understanding and transparency; developing user-friendly communication templates to maintain stakeholder involvement without causing fatigue. Learn more about the initiative here: https://holifoodproject.eu/citizen-and-societal/living-labs/
Climate change is driving the emergence of new food safety risks, but current risk communication strategies are not adequately tailored to inform and influence consumer behavior regarding these emerging risks.
Semi-structured interviews were conducted with consumers in Europe to understand people’s priorities for risk communication in relation to emerging food safety risks, and to understand if this differed from their preferences and priorities for communication about established risks.
It was concluded that emerging food safety risk communication needs to be developed and implemented to engage citizens in developing protective behaviours in relation to emerging food safety risks if specific behavioural changes are needed. At the same time, communication about adaptive and mitigatory actions is needed to reduce the impacts of emerging food safety risks associated with climate change.
The results were that communication campaigns can leverage motivational elements, such as positive framing, relevant narratives, and co-benefits including health and economic savings, to support behaviour adoption and sustained risk mitigation. Risk communication must be aligned with these behavioural drives, particularly for vulnerable populations such as children, pregnant women, and the elderly, who are more exposed to (emerging) food safety issues.
Proactive, transparent communication based on behavioural insights ensures that both individuals and organisations are better prepared to navigate the evolving landscape of food safety risks in a changing climate.
Unpredictable weather patterns, increased pest outbreaks, and soil degradation are increasing and affecting crop yields and food quality, making proactive risk management more important than ever. Additionally, food safety communication often lacks clarity, leaving both farmers and consumers uncertain about best practices.
A semi-structured interview with 80 European participants showed difficulty distinguishing emerging from existing food safety threats like antibiotic resistance and mycotoxins. While systemic risks linked to climate change were recognized, the lack of clear differentiation hampers prevention efforts.
Proactive measures, like early contaminant monitoring, reduced pesticide use, and efficient water management, can mitigate risks and support long-term profitability. Transparent, accessible communication builds consumer trust, with tools like QR codes and traffic-light labels seen as effective.
Participants valued clear, actionable messages. Collaboration among policymakers, industry, and farmers is essential to balance food safety with economic sustainability. Embracing innovation and sustainable practices will help ensure a safer, more resilient food system for the future.
Through enhancing early warning systems and analytical capabilities, we can proactively identify and mitigate threats, ensuring that the food we eat is safe and secure.
In this webinar, both EU-funded HOLiFOOD and FoodSafeR collaborated together to share their knowledge on food safety risks and why improving food safety risk analysis is crucial in safeguarding public health and maintaining consumer trust.
The webinar covered 3 topics that touches upon “Site Analytics”:
- Topic 1: Towards rapid on-site plant toxin detection
- Topic 2: Targeted methods for on-site testing of chemical hazards
- Topic 3: Targeted and untargeted analysis of plant and fungal metabolites.
For stakeholders in the food system and anyone concerned about food safety risks, this webinar offered insights and tools. You gain a deeper understanding of cutting-edge methods for detecting food hazards, ultimately enhancing consumer protection and contributing to a healthier, safer food system.
This webinar taught us that its best to:
- Utilise combined targeted and untargeted analysis approaches for more comprehensive safety screening
- Integrate these analytical capabilities into existing food safety management systems as early warning tools.
We are currently living in a fast-changing environment, and we are paying closer attention to what we eat. Keeping track of an ever-changing environment is critical for food safety and, by extension, human health. Even better is to be ahead of new dangers that may jeopardize food safety. As a result, the EU-funded projects FoodSafeR and HOLiFOOD focused on the early detection of new food safety issues through a webinar.
The webinar covered 4 topics on “System design for identification of emerging food safety risks”:
- Topic 1: FoodSafeR HUB
- Topic 2: Data Sharing Platform (architecture)
- Topic 3: Data Collection Platform (tooling)
- Topic 4: Federated Learning for access to food safety data
Main findings:
- AI & Big Data for Horizon Scanning
Researchers presented AI tools capable of scanning diverse data streams (scientific publications, media, reports) to flag emerging risks before traditional systems detect them. - Open Risk Identification Frameworks
The HOLiFOOD initiative promotes transparent, modular frameworks that can be adapted by food safety authorities, making advanced tech accessible and customizable. - Multi-Stakeholder Integration
A standout innovation is how the system encourages real-time input and validation from regulators, industry, and researchers—boosting trust and accuracy.
Watch the webinar to learn more about food safety risks and how a solid system design makes the difference.
https://holifoodproject.eu/lunch-webinar-on-system-design-for-identific…
In general we could preventive risk mitigation and protect the public health through a comprehensive early warning system.
A system design approach that combines artificial intelligence, machine learning, and expert knowledge to detect upcoming food safety hazards.
To tell you more about this, the 2 EU-funded initiatives (HOLiFOODproject and FoodSafeR) hosted a lunch webinar on "System Design for Identification of Emerging Food Safety Risks"
The webinar covered 4 topics on “System design for identification of emerging food safety risks”:
- Topic 1: FoodSafeR identification and characterization of drivers of change for emerging risks
- Topic 2: AI-driven prediction of emerging food safety risks: a holistic approach
- Topic 3: Emerging food risk detection through AI-enhanced weak signal mining
- Topic 4: Text mining models for emerging risk identification
Main findings:
- A privacy-preserving AI method that allows multiple stakeholders to train models on decentralized data enables cross-border collaboration in food safety without compromising proprietary or sensitive data.
- A modular online platform for food safety professionals to collaborate, share knowledge, and monitor trends. It empowers researchers and policymakers with a real-time pulse on emerging risks and best practices.
- Automated global food safety monitoring using web scraping, multilingual processing, and dashboarding offers a scalable, real-time source of structured data across languages and regions.
Watch the webinar to learn more: https://holifoodproject.eu/lunch-webinar-on-emerging-risk-identificatio…;
Contacts
Project email
Project coordinator
-
STICHTING WAGENINGEN RESEARCH
Project coordinator
Project partners
-
EUFIC
Project partner
-
AGROKNOW IKE
Project partner
-
ANSES
Project partner
-
APRE
Project partner
-
German Federal Institute for Risk Assessment
Project partner
-
Creme Global
Project partner
-
DIALOGIK
Project partner
-
Technical University of Denmark
Project partner
-
OnePlanet Research Center
Project partner
-
SECALIM
Project partner
-
University of Veterinary Medicine Budapest
Project partner
-
Newcastle University
Project partner
-
National Research Center (CNR)
Project partner
-
University of Vienna
Project partner
-
Laboratory of Food Microbiology of Wageningen University
Project partner
-
Queen's University Belfast
Project partner