An election analyst used ChatGPT to simulate the 2028 U.S. presidential race between Vice President JD Vance and California Governor Gavin Newsom, producing detailed state-by-state forecasts that went viral on YouTube.
The simulation, posted by the Election Time channel, walks through a complete Electoral College map filled in by poll-closing order, based entirely on the artificial intelligence model’s projections for the hypothetical general election matchup. The video has drawn fresh attention to speculation about the next presidential race.
What Happened
The host began by asking ChatGPT to select the nominees for both parties. On the Democratic side, the model selected California Governor Gavin Newsom, despite early polling showing him trailing former Vice President Kamala Harris by 15 points. “According to market odds on Kalshi, Newsom is the outright favorite with a 19% chance of winning his party’s primary ahead of a surging AOC and Jon Ossoff,” the host noted.
For the Republican nomination, ChatGPT chose Vice President JD Vance, described as the clear leader in polls. “According to Kalshi, the vice president has a 41% chance of being the next GOP nominee,” the host said, setting up the matchup between the two candidates.
The simulation then unfolded hour by hour, state by state. At 7 p.m. Eastern, polls close in six states and the District of Columbia. Indiana and Kentucky were called solidly for Vance, while South Carolina leaned Republican by 14 points. Georgia shifted to Vance by 4 percent—described as doubling former President Trump’s margin from 2024—while Virginia remained leans Democratic for Newsom by 3 points. After the first wave, Vance led 44 electoral votes to Newsom’s 16.
The 8 p.m. Eastern closing wave, covering 171 electoral votes across 16 states and D.C., produced significant developments. New Hampshire leaned Republican for Vance by 0.9 percent, marking the first GOP win there since George W. Bush in 2000. Pennsylvania leaned Republican, and Florida was locked in as likely Republican by 12 points. Newsom secured Massachusetts, Rhode Island, Maryland, D.C., Connecticut, Delaware, and Illinois. By the end of that hour, Vance held 178 electoral votes, needing fewer than 100 more to reach 270.
The Reaction
The 9 p.m. closings delivered what the host called a historic flip: Minnesota leaned Republican for Vance by 1.2 percent, marking the first Republican win there since Richard Nixon in 1972. Wisconsin and Michigan both leaned Republican by 2 to 4 points, while Arizona leaned Republican by 4 percent and Nevada locked in as lean Republican by 3 to 4 points.
The final results showed Vance winning 326 electoral votes and Newsom taking 212, according to ChatGPT’s simulation. The forecast included Republican victories in states traditionally Democratic—New Hampshire, Minnesota, and parts of Maine—while Newsom held the West Coast, the Northeast, and scattered Democratic strongholds.
What This Means
The host concluded that under ChatGPT’s scenario, Newsom’s performance would be the worst for any Democratic nominee since Michael Dukakis in 1988. The simulation represents one artificial intelligence model’s interpretation of electoral dynamics, demographic shifts, and hypothetical candidate appeal, though it carries no predictive authority and remains speculative in nature.
The viral video demonstrates growing public interest in computational approaches to election forecasting, even as the actual 2028 race remains more than two years away. Polling, market odds, and historical precedent continue to shape how analysts and the public contemplate future electoral outcomes.
