A new AI-driven simulation of the 2028 U.S. presidential election has gained attention on social media after the YouTube channel **Election Time** collaborated with **Grok AI**, the artificial-intelligence platform created by Elon Musk’s company xAI.
The video presents a hypothetical Electoral College scenario featuring Kamala Harris against JD Vance. Using historical voting patterns, demographic trends, polling data, and betting markets, the model simulates potential outcomes and explains how different strategic factors might shape the race. Although the creators emphasize that it is not a prediction, the project has drawn interest as an example of how AI modeling can explore possible political scenarios for the 2028 United States presidential election.
In the simulated Democratic primary, Grok AI places Kamala Harris in the lead with about 32 percent support. She is followed by Gavin Newsom at roughly 23.8 percent, while Pete Buttigieg sits near 10 percent. Other contenders include Alexandria Ocasio-Cortez and Josh Shapiro, illustrating a fragmented early field. The simulation also suggests Harris’s chances of entering the race have risen sharply—from about 11.2 percent months earlier to 56 percent—indicating a potential political comeback after her loss to Donald Trump in the 2024 United States presidential election. Factors such as fundraising ability, name recognition, and media attention are included in the model to explain her early advantage.
On the Republican side, the AI projects JD Vance as the clear frontrunner with about 49.2 percent support in early polling. He significantly outpaces competitors like Donald Trump Jr., who is modeled at around 20.2 percent, while Marco Rubio and Ron DeSantis trail further behind. The simulation assigns Vance roughly a 46 percent probability of becoming the Republican nominee, compared with Rubio’s 18 percent. The analysis highlights how incumbency, party loyalty, and recent Republican gains in swing states could strengthen Vance’s position, though it also notes that unexpected political developments could still reshape the race.
When constructing the Electoral College map, the model first allocates “solid” states—those projected to be decided by margins greater than 15 points. Vance secures many of these across the Midwest, Mountain West, and Deep South, including traditionally Republican states such as Utah, Idaho, Wyoming, Oklahoma, and Alabama. Notably, Ohio is classified as solid Republican, reflecting its continued shift to the right after recent election cycles. Harris maintains a group of solid Democratic states including California, Washington, Massachusetts, Maryland, and the District of Columbia. At this stage of the map, Vance leads with 139 electoral votes compared to Harris’s 108.
The simulation then adds “likely” states, where margins fall between 5 and 15 points. Vance expands his lead by winning states such as Florida, Texas, Arizona, Iowa, and North Carolina, illustrating continued Republican strength in the Sun Belt and parts of the Upper Midwest. Harris’s likely states include Oregon, Colorado, Illinois, New York, and Virginia. However, the model flags Illinois and New York as potentially weaker than in previous cycles due to reduced projected margins. After combining solid and likely states, Vance reaches 246 electoral votes while Harris holds 212.
Finally, the simulation examines “lean” and “tilt” states that could ultimately decide the election. Nevada and Georgia are categorized as lean Republican, while the key Midwest battleground trio—Wisconsin, Michigan, and Pennsylvania—also narrowly favor Vance. Harris holds lean advantages in New Jersey and Nebraska’s second congressional district, but the model places Minnesota and New Hampshire in the extremely competitive “tilt Republican” category. In the final projection, JD Vance wins 326 electoral votes, carrying all states won by Trump in 2024 and narrowly flipping Minnesota and New Hampshire, while Harris’s support remains concentrated along the West Coast and parts of the Northeast.
