Background <p>Anticancer peptides (ACPs) are short bioactive molecules capable of inhibiting cancer cell proliferation, migration, and invasion while selectively targeting malignant cells through various mechanisms, such as membrane disruption, apoptosis induction, and immune modulation. They are widely distributed across all kingdoms of life, including bacteria, fungi, plants, and animals (vertebrates and invertebrates), as well as in viruses such as bacteriophages and eukaryotic viruses. They can also be synthesized artificially. ACPs have gained attention due to their cationic nature, high selectivity, and relatively small size compared to conventional biological therapeutics (i.e., antibodies, proteins).</p> Objective <p>This review aims to provide an updated overview of ACPs, including their diverse natural and synthetic sources, principal mechanisms of action, their progression through clinical development, key challenges, emerging computational and AI-based discovery strategies, and their therapeutic potential in cancer treatment.</p> Methods <p>A comprehensive analysis of published literature was conducted to summarize ACPs derived from diverse natural and synthetic sources. Publicly available ACP and peptide-drug databases, computational prediction tools, and emerging machine learning and deep learning approaches were also examined to highlight current capabilities and future directions.</p> Results <p>Anticancer peptides from diverse biological sources exhibit multiple, often overlapping mechanisms of action, including membrane disruption, apoptosis induction, DNA damage, inhibition of microtubule assembly, suppression of migration and metastasis, and immune modulation. Several ACPs and peptide-based agents, such as LTX-315, Motixafortide, Balixafortide, CIGB-300, Plitidepsin, and various peptide vaccines, have progressed to Phase I–III clinical trials, with some receiving FDA and EMA regulatory approval. Available databases and computational tools support ACP identification and design, while recent machine learning, deep learning, and AI-aided approaches show significant potential for improving peptide activity, stability, selectivity, and overcoming therapeutic limitations, such as rapid degradation, toxicity, and poor/inefficient delivery.</p> Conclusion <p>Anticancer peptides represent a promising class of therapeutic agents. The integration of experimental research with computational, including machine learning and AI-driven approaches, may enhance their design, improve efficacy and safety, and accelerate their clinical translation toward next-generation cancer therapeutics.</p>