The potential natural vegetation is the most stable and mature climax vegetation type that achieves a final, balanced state without human interference under a given climate condition, and it can be used to reflect the dominant growing trend for local vegetation. Therefore, the investigation of the potential vegetation on a regional scale improves understanding of the interaction mechanism between vegetation and the climate system, and provides useful references for the restoration and construction of regional vegetation. With the support of Geographic Information System (GIS) technology, the meteorological data in China under the scenarios A1B, A2, and B1 between 2070 and 2099 were obtained, and the Comprehensive Sequential Classification System (CSCS) was applied. The spatial distribution of potential vegetation under different climate change scenarios was simulated, and the sensitivity of future potential vegetation under different scenarios was analyzed based on the Comprehensive Sequential Classification System. The results show that (i) the spatial patterns of China's potential vegetation are generally similar under the three different climate change scenarios. The horizontal and vertical zonality of potential vegetation in China was identified and was very similar to the distribution of the existing natural surface vegetation. However, there are still spatial differences in some areas, such as the Qinghai-Tibet Plateau where there were less potential vegetation types. Detailed differences also exist in type and area of the potential vegetation in China under the different scenarios. A total of forty potential vegetation types were simulated under the A1B scenario, of which the "cold temperate perhumid taiga forest type (IIF)" showed the widest distribution with an area of 8.26 × 10⁵ km² at 8.61% and the "tropical arid tropical desert brush type (VIIB)" had the smallest area at 1.35 × 10⁵ km² (0.01%). Forty-one potential vegetation types were found under the A2 scenario, of which the "frigid perhumid rain tundra, alpine meadow type (IF)" had the widest distribution with an area of 9.06 × 10⁵ km² (9.44%) and the "frigid extra-arid frigid desert, alpine desert type (IA)" has a limited area of 6.90 × 10 km² (less than 0.01%). Thirty-nine types were included under the B1 scenario, of which "frigid perhumid rain tundra, alpine meadow type (IF)" was the largest at 1.40 × 10⁶ km² (14.59%) and "tropical arid tropical desert brush type (VIIB)" had only a small area of 4.03 × 10² km² (0.004%). (ii) The "tropical extra-arid tropical desert type (VIIA)" does not exist in China under all three scenarios (A1B, A2, and B1) in the 20^th and 21^st centuries. This suggests that China's climate conditions are not suitable for the formation and development of the "tropical extra-arid tropical desert type (VIIA)." (iii) A comparison of the potential vegetation in the same region under the three different climate change scenarios showed that the sensitive zones account for 64.10% of China's total territory. The proportion of the sensitive zones in Northwest China, North China, South China, and Tibetan Plateau region were 68.20%, 70.82%, 49.94%, and 66.59%, respectively. However, the insensitive zones account for 35.90% of the total land area, and the potential vegetation distributed in these insensitive areas was hard to change with climate change.