Weeks before U.S. forces captured Nicolás Maduro in Venezuela, Donald Trump reportedly spent hours in the Oval Office asking Elon Musk’s artificial-intelligence chatbot questions about his presidency, his legacy — and what might happen if Maduro were removed from power.
One of those questions now stands out.
Trump asked Grok how Venezuelans would react if the United States captured Maduro.
According to a new TIME report, the chatbot responded that Maduro was deeply unpopular and that many Venezuelans would likely celebrate his removal.
The exchange took place during a private December 2025 meeting between Trump and Musk.
An administration official who was present told TIME that Trump spent hours experimenting with Grok and came away impressed by what it could do.
Then, the following month, Trump ordered the operation that seized Maduro.
The sequence is remarkable.
But it is important not to overstate what the reporting establishes.
TIME does not report that Grok made the decision for Trump.
It does not say the chatbot was the sole reason the operation went forward.
And it does not establish that Trump treated an AI response as a substitute for military or intelligence advice.
What it does reveal is that Trump was using an AI chatbot to explore a highly consequential foreign-policy scenario shortly before acting on essentially that scenario.
At the time of the Musk meeting, tensions with Venezuela were already extremely high.
The United States had been carrying out military strikes against vessels Trump said were involved in drug trafficking.
Trump was simultaneously considering increasingly aggressive action against Maduro’s government.
During the Grok session, Trump reportedly asked how ordinary Venezuelans would respond if Maduro were captured.
The chatbot characterized Maduro as repressive and unpopular and suggested that many people would welcome his downfall.
After the January operation, public celebrations did take place in parts of Venezuela.
According to TIME, that apparent match between Grok’s prediction and subsequent events left Trump believing the chatbot was “ingenious.”
The episode offers an unusual glimpse into how Trump has begun interacting with artificial intelligence during his second term.
He has become increasingly enthusiastic about the technology.
He has surrounded himself with leading AI executives.
And he has repeatedly argued that the United States must move faster than China in developing advanced systems, even as lawmakers and researchers raise concerns about safety and oversight.
But the Venezuela episode is especially significant because it intersects with decisions involving military force.
Maduro’s capture became one of the defining events of Trump’s second year back in office.
U.S. forces removed the Venezuelan leader in January, and Maduro was subsequently taken into American custody to face federal charges.
Venezuela was then placed under an interim government led by Delcy Rodríguez, with Washington exerting considerable influence over the country’s political and economic direction.
The rapid operation appears to have had a broader effect on Trump’s thinking.
TIME reports that the successful raid reinforced his belief that bold military action could succeed where more cautious approaches had failed.
That confidence then became relevant as the administration moved toward war with Iran.
Trump launched U.S. attacks on Iran on Feb. 28.
He initially described the conflict as something that could be resolved relatively quickly.
Instead, the war stretched into its seventh month, depleted certain U.S. munitions, increased energy prices and generated significant military costs and casualties.
TIME describes the Iran decision as a major turning point in Trump’s second term.
The magazine’s reporting argues that the Venezuela operation had strengthened Trump’s confidence in decisive uses of force.
Iran proved far more complicated.
By late September, Trump was still publicly saying the United States would achieve victory “very soon,” while diplomatic efforts continued without a lasting resolution.
That distinction is important.
There is evidence that the success of Venezuela influenced Trump’s broader strategic confidence.
There is evidence that he consulted Grok about the likely reaction to Maduro’s capture.
But there is not evidence that Grok “talked Trump into” either the Venezuela operation or the war with Iran.
Those decisions involved far larger national-security processes.
Still, the fact that a sitting president reportedly used a consumer AI chatbot to explore a potential military operation is striking on its own.
AI systems can produce fluent, confident answers.
But they are not intelligence agencies.
They do not independently verify every claim they generate.
Their responses can reflect incomplete data, flawed assumptions or errors embedded in training material.
That makes the role of such tools in high-stakes government decision-making an increasingly important question.
Trump has shown little interest in slowing the technology’s expansion.
At a White House meeting on Sept. 29, he brought together major technology executives including leaders from Meta, Google, Nvidia, OpenAI and other firms.
Trump and six technology companies signed what he described as a “constitution” for AI safety.
The agreement is voluntary rather than a binding federal regulatory framework.
Trump has also pushed officials to use the phrase “super intelligence” rather than simply artificial intelligence.
He has announced plans for an “AI Force,” or “SI Force,” made up of industry leaders advising his administration.
Asked by TIME what the group would actually do, Trump offered few specifics beyond saying it would be “super intelligent.”
His administration this week also launched America.gov, an AI-powered portal intended to help Americans navigate federal information and services.
The platform is designed to answer questions using government sources and eventually simplify tasks involving passports, Medicare, employment and other federal programs.
Trump’s embrace of AI therefore extends far beyond novelty images or social-media experiments.
The technology is moving closer to the center of his administration.
What the Grok episode reveals is how personal that enthusiasm has become.
In December, Trump sat in the Oval Office with Elon Musk and asked an AI chatbot what might happen if the United States captured the leader of another country.
Weeks later, U.S. forces did exactly that.
The available evidence does not show that Grok made the decision.
But it does show that artificial intelligence had become part of the conversation before one of the most consequential foreign-policy actions of Trump’s second term.
And according to officials who spoke to TIME, the outcome only made Trump trust the technology more.
