A recent AI-generated prediction about the 2028 U.S. presidential election has stirred conversation online, not because it offers certainty, but because it presents a structured, data-driven guess at how the political landscape might evolve in the years following Donald Trump’s current term. With several years still remaining before voters head to the polls, the idea of forecasting a winner now may seem premature—but that hasn’t stopped people from paying attention to what the model suggests.

The simulation, reportedly based on trends, polling behavior, and current political positioning, focused heavily on two prominent Republican figures: JD Vance and Marco Rubio. Both are seen as potential successors within the same political orbit, but the AI drew a distinction between their current roles that could significantly impact how voters perceive them in the future. According to the model, proximity to Trump’s presidency could either be an advantage or a liability depending on how the next few years unfold.

For Vance, serving as Vice President places him directly alongside Trump in both achievements and controversies. The AI suggested that this close association could make it difficult for him to distance himself from any unpopular decisions or political turbulence that may arise. In contrast, Rubio’s role as Secretary of State might offer him a different narrative—one focused more on foreign policy than domestic issues—giving him potential flexibility in how he positions himself during a campaign.

Interestingly, despite this perceived advantage for Rubio, the model still identified Vance as the early favorite within the Republican field. The reasoning was straightforward: incumbency-adjacent visibility often translates into stronger recognition and loyalty among voters. However, the AI emphasized that this lead is fragile. If Vance were to face setbacks—whether political, economic, or reputational—the balance could shift quickly, creating an opening for Rubio to gain ground.

One of the more striking elements of the prediction involves Trump himself. The model suggested that his influence could remain a decisive factor in shaping the Republican primary outcome. An endorsement from Trump, according to the simulation, could effectively determine the nominee, reflecting the continued weight of his support within the party. This highlights a broader theme: even after leaving office, Trump’s political presence may continue to shape outcomes in significant ways.

On the Democratic side, the AI pointed to Gavin Newsom as the most likely nominee and, notably, as a potential overall winner in the general election. The reasoning here centered less on individual policy and more on voter psychology. The model introduced the concept of “political exhaustion,” suggesting that after years of intense and often divisive political cycles, voters might lean toward a candidate perceived as representing a shift in tone or direction.

According to the simulation, such a scenario could lead to a narrowly decided election, with Newsom edging out a Republican opponent in a closely contested race. However, it’s important to understand that this outcome is not a prediction in the traditional sense—it’s a scenario based on current data and assumptions, all of which can change dramatically over time.

That’s really the key point behind all of this: AI models don’t know the future. They analyze patterns, weigh probabilities, and generate outcomes based on available information. But politics, perhaps more than any other field, is shaped by unpredictable events—economic shifts, global crises, candidate decisions, and public sentiment that can change almost overnight.

There’s also a tendency to treat AI outputs as more definitive than they actually are. In reality, they are best understood as tools for exploring possibilities rather than declaring outcomes. The closer an election gets, the more accurate data becomes, but even then, surprises are common. Looking three years ahead, any projection—AI-generated or otherwise—should be taken with caution.

What makes this particular prediction “eye-opening” isn’t that it claims to know who will win, but that it highlights the factors likely to matter: association with current leadership, the ability to shape a personal narrative, and the broader mood of the electorate. These are variables that analysts have long considered, now simply packaged through a different lens.

So while the idea of an AI forecasting the next U.S. president might sound dramatic, the reality is more grounded. It’s not about certainty—it’s about perspective. And in a political environment as fluid as the United States, perspective can shift just as quickly as the headlines themselves.