Trump consulted Grok for hours before US captured Venezuela's Maduro, Time reports

A secret meeting with a chatbot

According to Time magazine, President Donald Trump held a secret meeting with Elon Musk in December 2025, roughly one month before the United States invaded Venezuela and captured its president, Nicolás Maduro. The meeting came about seven months after Musk exited his role in the Trump administration at the Department of Government Efficiency.

TechCrunch, summarizing Time's reporting, says Trump spent hours in conversation with Musk's Grok chatbot during the session. Among other things, a source told Time, he asked how Venezuelans would react if their president were captured.

What led up to the invasion

The consultation followed months of escalation. In September 2025, Trump began ordering US military strikes on Venezuelan boats that he alleged were involved in drug trafficking. As The Atlantic reported at the time, users were already turning to Grok with questions about those strikes and about Venezuela's political situation.

According to Time, Grok's answer to Trump characterized Maduro as a "deeply unpopular dictator" and suggested that many Venezuelans would likely celebrate his downfall. When the US invaded on January 3, 2026, and such celebrations took place, the source told Time that Trump came away thinking Grok was "ingenious."

Grok's growing footprint at the Pentagon

The episode fits into a wider pattern of Grok embedding itself in US defense work. TechCrunch notes that in June, the Pentagon's head of AI said the military had used Gov Grok, a government-facing version of the chatbot, to deploy and strike targets during the Iran War.

Grok is not the Department of Defense's only AI supplier. TechCrunch reports that OpenAI also holds an agreement with the department, and that Anthropic has been in extended discussions over how its models could be applied to military intelligence and modern warfare. Grok's position may strengthen further under the current administration: earlier this week, the Pentagon announced that Musk and Palmer Luckey, founder of Anduril, will co-lead a study on how advanced technology is, and can be, used on battlefields.

Why it matters

The reporting, which TechCrunch surfaces from Time's investigation, raises a question that the AI industry has so far mostly discussed in hypothetical terms: what happens when a head of state treats a consumer chatbot as a policy oracle? Grok's assessment of Venezuelan public sentiment happened to match what followed the invasion, but a language model's prediction about how a population responds to regime change is not intelligence in the traditional sense. According to the sources cited, there is no indication the answer was verified, stress-tested, or weighed against dissenting analysis before it reached a president weighing military action.

The story also blurs the boundary between one company's commercial product and national security decision-making. The same firm whose chatbot reportedly reinforced a decision to invade now supplies a government-grade model used in strike operations, and its owner has been appointed to co-lead a Pentagon study on battlefield technology. For readers tracking AI governance, this is a concrete case of consumer chatbots migrating into consequential government roles faster than oversight, procurement norms, or accountability mechanisms have developed around them.

Neither the White House nor xAI has publicly responded to the specifics of Time's account, according to the material available. If the reporting holds, it marks one of the clearest examples yet of a generative AI system directly influencing a major geopolitical decision.