Cosmology has entered an era of unprecedented precision, yet increasing accuracy has revealed cracks in the standard $\Lambda$CDM paradigm. Although the model remains highly successful when confronted with individual datasets, joint analyses expose a network of tensions involving the Hubble constant, CMB lensing, curvature, neutrino masses, and the nature of dark energy.
In this contribution to the 3rd General Meeting of the COST Action COSMIC WISPers (CA21106), within the context of Working Group~2, we critically assess these discrepancies, emphasizing the role of model assumptions, parameter degeneracies, and dataset consistency. We review proposed early- and late-time solutions, discuss how recent DESI BAO results alter the viability of late-time extensions, and explore interacting dark-sector scenarios.
Our analysis highlights the need for caution in interpreting cosmological measurements and underscores the importance of internal consistency among cosmological probes before claiming percent-level accuracy or invoking new physics.