Donald Trump said on Tuesday that top executives in the technology sector had signed a new pledge aimed at improving the “self-regulation” of their companies’ artificial intelligence models, arguing that a “light-touch” approach is the best way to balance safety concerns with the desire to accelerate the development of a technology he has dubbed “Super Intelligence.”
Flanked by technology industry executives as he spoke to reporters in the East Room of the White House and later outside the West Wing, Trump praised the CEOs who attended a White House luncheon, while dismissing concerns about the risks of rapidly advancing artificial intelligence technology.
ΣΧΕΤΙΚΑ ΑΡΘΡΑ
“I see a tremendous amount of self-regulation, and they understand that they need to self-regulate,” he told reporters. “These models are very complex, and, I repeat, I believe they will be used for the common good.”
The president then posted the agreement on Truth Social, which detailed four measures that all companies agreed to take:
- Implement internal controls to ensure that their models comply with appropriate cybersecurity standards.
- Authorize an internal team to ensure that controls, monitoring, and detection are functioning as intended.
- Work with an external firm to conduct an independent assessment of the models.
- Appoint an “independent board committee” to receive these reports and ensure that any issues are addressed.
In addition to Trump, the post showed six signatures on the document: Sundar Pichai of Google, Dario Amodei of Anthropic, Mark Zuckerberg of Meta, Jensen Huang of Nvidia, Greg Brockman of OpenAI, and Elon Musk of X.
According to CNNi, the meeting took place following a series of incidents that raised widespread safety concerns regarding artificial intelligence, including the uncontrolled behavior of AI agents and attempts to breach federal databases.
Several officials in the artificial intelligence industry, including some CEOs who are leading the development of increasingly sophisticated AI models, have even warned of catastrophic consequences should they lose control of their technology.
However, the meeting did not seem to dampen Trump’s enthusiasm for the models he himself calls “Superintelligence,” but rather reinforced his belief that the artificial intelligence industry is capable of self-regulation without additional safety regulations. He described the new commitment as “morally binding.”
Trump, who stated that he would soon sign an executive order renaming “artificial intelligence” to “superintelligence,” insisted that the government cannot slow down the development of artificial intelligence for fear of falling behind China and other developed countries.
“This is an incredible development for our country,” he said. “Whoever masters superintelligence will win.”
What We Know About the Meeting
Among those in attendance on Tuesday were Amodei, Zuckerberg, Huang, Brockman, Musk, Pichai, Alex Karp of Palantir, and Jeff Bezos of Amazon, along with several members of his Cabinet and House Speaker Mike Johnson. Sources familiar with the meeting described it as an effort to coordinate between the administration and top industry executives, amid growing security concerns related to artificial intelligence models.
Brandon Daniels, CEO of Exiger, who also attended the meeting, stated that technology industry leaders focused on an idea proposed by Huang and Zuckerberg, according to which the group needed to “align itself with a set of principles and standards that would take precedence—not just individual ones, not just everyone for themselves — a set of principles and standards that can be revised, that may be subject, you know, perhaps to future executive orders or future regulations.”
He described it as “a pact that allows us to implement these standards, to implement governance, and then have them audited and inspected by third parties.”
“We agree that, you know, the existence of central standardization bodies, as well as audits and inspections, and also internal procedures through which we conduct and develop artificial intelligence, are things we all believe we need to have, and we consider them to be things that could be subject to central oversight,” Daniels added.
Daniels described the meeting as “truly friendly” and noted that the leaders, who had not yet been in contact with one another, exchanged cell phone numbers. He praised the president for bringing the team together, describing his actions as “revolutionary.”
In recent weeks, Trump has repeatedly stated that he believes the term “artificial intelligence” is incorrect, since, in his view, this technology should be called “superintelligence.” He said he even raised the issue of renaming it with Chinese leader Xi Jinping, who “seemed to like” the idea. (In his own remarks, Xi did not refer to it as “superintelligence.”) The State Department had already instructed diplomats in its Office of International Organizations to use the term “superintelligence,” according to the Associated Press.
Source: CNN.gr
