Climate change has become a significant systemic risk in the global financial landscape, yet traditional asset pricing models such as the Capital Asset Pricing Model (CAPM) and the Fama-French Three-Factor Model have not explicitly integrated this risk. A report by the BlackRock Investment Institute (2020) suggests a 5–10% valuation reduction for companies with high climate risk exposure, while the Swiss Re Institute (2021) projects an economic impact of up to 18% of global GDP by 2050. This gap highlights the urgent need to revise asset pricing models to reflect the reality of climate risk. This study aims to synthesize approaches to integrating climate risk (physical and transition) into asset pricing models, identify methodological challenges in its measurement and representation, and evaluate the validity of climate risk metrics such as ESG scores and climate beta. Using a narrative review approach, this study analyzes literature from Scopus, Web of Science, JSTOR, and ScienceDirect databases (2010–2025). The search strategy involved keywords such as “climate risk”, “asset pricing”, “ESG”, and “machine learning”. Thematic analysis was applied to identify patterns in theoretical approaches, climate variables, adopted models, and empirical results. Findings suggest that climate risk integration can be achieved through the Climate Beta concept, ESG-based models, portfolio sorting based on carbon exposure, and hybrid models with machine learning. Empirical evidence suggests that carbon-intensive sectors experience lower stock returns and increased volatility, although methodological inconsistencies remain a challenge. Integrating climate risk into traditional asset pricing models, particularly through the development of adaptive multifactor models, has the potential to improve return prediction and risk assessment. More robust methodologies, standardized environmental data, and multidisciplinary collaboration are needed to produce valid and applicable models.