EFSA Outlines Scientific Needs for Regulatory Preparedness, Risk Assessment

On September 29, the European Food Safety Authority (EFSA) published a paper outlining the scientific capabilities and investments that will be necessary to keep pace with food system innovation, emerging risks, and changing EU policy, while enabling effective and efficient risk assessment.
The paper identified scientific needs across six interconnected pathways:
- Innovation and competitiveness
- Methodological and technical advances
- Sustainability and One Health
- Preparedness and rapid response
- Risk communication and social science
- Artificial intelligence (AI) and data.
The paper is intended to inform EFSA's forthcoming Strategy 2034 and discussions at the agency’s Safe Food, One Future Conference, taking place in fall 2027 in Parma, Italy.
Regulatory Challenges Arising from Food Innovation
EFSA highlighted several areas in which innovation is creating new regulatory science challenges, including cell culture-derived foods, precision fermentation, plant molecular farming, microbiome-based products, biopesticides, bacteriophages, new genomic techniques (NGTs), circular food and feed sources, novel food contact materials, and emerging agricultural systems.
Specific challenges described in the paper included evaluating cultured animal cell products, innovative feed additives, biopesticides, and novel recycled food contact materials and recycling technologies.
EFSA also pointed to scientific gaps involving allergenicity and protein safety, microbiomes, environmental risk assessment, and potential health risks associated with micro- and nanoplastics.
The agency said fit-for-purpose risk assessment methods will be needed to address novel products, advanced materials, emerging evidence streams, and scenarios involving multiple hazards.
Calls for Advances in NAMs, Exposure Science, and Probabilistic Risk Assessment
EFSA identified continued research and investment needs related to new approach methodologies (NAMs), which could support the transition away from animal testing. Although progress has been made, EFSA said single alternative methods are insufficient for comprehensive hazard assessment. Further work is needed on integrated testing batteries, toxicokinetic methods, quantitative adverse outcome pathways (qAOPs), omics, and the integration of new evidence and data types into regulatory risk assessment.
Looking for quick answers on food safety topics?
Try Ask FSM, our new smart AI search tool.
Ask FSM →
The paper also called for advances in exposure science, including better incorporation of human biomonitoring, aggregate exposure, exposome information, lifetime exposure modeling, and non-targeted screening.
Additionally, EFSA said probabilistic approaches could improve assessments of combined exposure to multiple chemicals and other complex exposure scenarios for which threshold-based approaches alone are insufficient.
One Health, Surveillance, and Climate-Related Food Safety Risks
EFSA identified One Health cooperation as another important component of future regulatory readiness. Areas requiring further investment included antimicrobial resistance (AMR), microbiomes, biomarkers of effect, animal welfare, vector-borne diseases, integrated monitoring and surveillance, environmental risk assessment, and climate change.
For example, EFSA called for expanded interoperable surveillance systems capable of integrating information across sectors and delivering information rapidly enough to facilitate alerts. The Authority also identified a need to better quantify climate change's effects on food safety, including impacts on aflatoxins and the spread of vector-borne diseases.
EFSA said future preparedness should increasingly move from reactive crisis response toward an anticipatory, systems-based model. Proposed capabilities included AI-enabled horizon scanning, machine learning and natural language processing, early warning systems, scenario-based foresight, predictive modeling, and digital twins for simulating food system vulnerabilities.
EFSA Envisions Human-Led, AI-Enabled Risk Assessment
AI and data were featured prominently in EFSA's vision for future regulatory science. The agency said it should move from largely document-based risk assessment toward a human-led, AI-enabled model in which scientific authority remains with EFSA experts, panels, staff, and accountable governance structures.
Under such a model, AI could support tasks including problem formulation, literature searches, evidence retrieval and screening, data extraction, critical appraisal, evidence synthesis, uncertainty analysis, dossier review, drafting, and quality checks. EFSA stressed that AI would support, rather than replace, scientific judgment.
EFSA also envisioned moving beyond PDF-based scientific opinions toward structured, machine-readable risk assessment packages containing evidence, data and model references, uncertainty information, traceability links, human review points, and reusable outputs.
The agency concluded that maintaining scientific excellence and regulatory readiness will require sustained research and innovation in methods, data infrastructure, specialist expertise, capacity building, partnerships, and research engagement. According to EFSA, integrating these capabilities could enable faster and more anticipatory scientific advice while maintaining scientific rigor, transparency, and public trust.









