Almost every human eats and drinks a considerable daily amount of food additives (FAs), accumulating towards 3.6–4.5 kg/year [1]. These additives are used to improve the foods’ taste, texture, aesthetic, and shelf life, representing a wide range of different chemicals with various properties. While commonly consumed worldwide, these compounds are increasingly attracting concerns about their potential impacts on human and environmental health.Scientific and public debates on FAs’ safety arose in the 1970s regarding the alleged neurobehavioral effects of some food additives [2]. In the 2000s, the so-called “Southampton study” again stirred the argument with the demonstration that consumption of FA mixtures may relate to hyperactivity in children [3], leading to long scrutiny of the infamous “Southampton Six” (Tartrazine, Quinoline Yellow, Sunset Yellow, Azorubine, Ponceau 4R, and Allura Red) by both scientists and legislators. Another research, the “Liverpool study”, showed that FA mixes might synergistically affect the viability and differentiation of mice NB2 neuroblastoma cells [4], adding three more FA suspects: Brilliant Blue, Monosodium Glutamate, and Aspartame. Since then, there have been various studies on the potential health effects of food additives, both individually and in mixtures [5,6,7].The rise in safety and toxicity studies has provided legislators with vast shreds of evidence to frequently and scientifically update their FA policies. However, there is a wide mismatch among policies worldwide on the safety level of each additive, largely due to the different rates of scientific updates, as well as to the different viewpoints of weighing evidence—for instance, the differences in “acceptable daily intakes” (ADIs) issued by the two most notable regulators: the European Food Safety Authority (EFSA) and the Joint FAO/WHO Expert Committee on Food Additives (JECFA) (Supplementary Table S1). Nevertheless, there is always the need for more comprehensive and highly reliable research into the different aspects of FA safety and potential toxicity, and any new piece of evidence on this topic is valuable for policymakers to reevaluate the substances [8].On the other hand, FAs are also listed among emerging water contaminants in all aquatic systems, from sewage to the ocean [9,10,11]. While multiple efforts are put into removing the pollutants, research into these compounds’ potential impacts on aquatic organisms is also required. Indeed, some of these compounds, such as Carmine and Sucralose, have already been shown to be aquatoxic, threatening global water environments [12,13].The need for more studies into FAs’ potential effects on both human and environmental health is in line with the One Health concept. It advocates the use of models that can represent both aquatic ecosystems and humans. These models should ideally be in vivo vertebrates, representing the complexity of an entire organism and maximizing the chance to capture unexpected outcomes. However, the recent trend of applying the 3R principle also requires minimizing the use of animals. Therefore, zebrafish embryos are a perfect candidate for this task: Firstly, despite being a complete lifeform, the zebrafish embryos up to the free feeding stage (120 h post-fertilization—hpf) are not legally recognized as animals in the EU [14], thus totally complying with the 3R. Secondly, the zebrafish is an aquatic vertebrate whose genome shares 70% orthologous genes with humans [15], hence representing both environment and human health. Thirdly, the zebrafish’s rapid embryogenesis allows observation and recapitulation of multiple targets and processes occurring during early development, which can be easily observed through the transparent chorion. Additionally, the fish’s high fecundity and low maintenance cost offer the prospects for developing high-throughput assays [16,17,18]. These advantages have made the zebrafish embryotoxicity test (ZET) an increasingly recognized tool in chemical safety screening for both environmental and biomedical applications [18,19,20,21,22].Over the years, the zebrafish embryotoxicological toolbox has been supplemented with various advanced techniques, such as transgenic reporter lines or automated phenotyping [23,24,25] to increase experimental throughput and simplify training, or-omics tools such as RNA-Seq [22,26] that enable researchers to explore the mechanisms involved in a chemical’s bioactivity. These methods, while screening for chemicals’ toxicity, often mainly focus on preset endpoints, such as lethality, simple morphological defects, or expression of a reporter gene. However, one big advantage of the embryonic zebrafish model is that specific phenotypes induced by a chemical can give hints to the underlying biological process, which is extremely important when it comes to the safety assessment of chemicals. Following up on unexpected phenotypes observed in zebrafish embryos may serve as the starting point for mechanistic studies on toxico-/pharmacology [22,26,27,28].In this study, we employed the zebrafish embryos as the model system to investigate the potential biological effects of controversial food additives, selected from the “Liverpool” and “Southampton” studies [3,4]. Thereby, we also demonstrate morphological phenotyping as an effective tool in suggesting chemicals’ mode of action involved.