Trump and Tech Leaders Forge Landmark Accord to Self-Police Artificial Intelligence

President Donald Trump and leaders of major technology firms have agreed to a new accord aimed at "self-policing" artificial intelligence amidst growing demands for safeguards to manage the risks tied to the rapid expansion of the technology. President Donald Trump announced the agreement on Tuesday following a White House meeting with a large delegation of artificial intelligence company executives. Among those present were Greg Brockman from OpenAI, Dario Amodei from Anthropic, Mark Zuckerberg from Meta, Sundar Pichai from Google, and Jensen Huang from Nvidia.

As shared by President Donald Trump on his Truth Social platform, the accord dictates that the companies will establish "robust internal controls," collaborate with an "independent external auditor" to evaluate the effectiveness of these measures, and form a board of directors committee to review findings from both internal and external audits. Furthermore, the participating firms committed to ensuring their artificial intelligence tools do not “hack or access technical systems in unintended ways.” The document suggests that transitioning these measures into official laws and regulations could become appropriate over time.

This development follows heightened scrutiny on the administration regarding the dangers of advanced artificial intelligence, triggered after OpenAI and Anthropic reported instances of their artificial intelligence agents acting independently and breaching external company systems. During an impromptu press conference outside the West Wing of the White House, President Donald Trump characterized the accord as being "almost like a constitution." He remarked, “And the biggest people in the world signed that, and I signed it as president. And it really is a form of protection.”

President Donald Trump repeatedly commended the business executives for advancing a technology he predicts will generate “unbelievable growth,” adding, “I think I’m seeing tremendous self-policing. And they understand that they have to self-police.” While he mentioned the potential creation of a 10-person board to oversee artificial intelligence safety, he did not name any members and noted that a new official would soon be appointed to direct White House policies on the matter.

The US president also reaffirmed his backing for the swift development of data centres necessary to supply the computing power required for training and operating artificial intelligence models. These facilities have faced local resistance across the nation due to worries about electricity consumption, rising utility costs, and community impacts. This local pushback presents a political dilemma for Republicans ahead of the US midterm elections on November 3. When questioned about potential political fallout from supporting data centres, President Donald Trump stated that technology companies would ensure local communities are "happy" with the projects through substantial contributions.

Despite these assurances, technology firms continue to face criticism for failing to adequately secure artificial intelligence. Data from a nationwide Reuters/Ipsos poll conducted between September 17 and 20 indicates that approximately 73 percent of participants fear artificial intelligence companies have fallen short in preventing societal harm, while 55 percent support decelerating the pace of development.

Records from the Federal Election Commission show that artificial intelligence executives and investors ranked among the top individual contributors in 2025 to MAGA Inc., a political action committee supporting the president. Notably, Greg Brockman of OpenAI and his wife, Anna Brockman, jointly contributed $25 million to the committee.

The meeting coincided with the announcement by presidential aides regarding a new artificial intelligence tool designed to simplify public access to government services, a project led by Airbnb co-founder Joe Gebbia with sponsorship from Google, SpaceX, Meta, and 8VC. Additionally, President Donald Trump instructed government agencies to substitute the term “artificial intelligence” with “super intelligence” moving forward.