Trump, Tech CEOs Sign Voluntary AI Safety Pact
US President Donald Trump and executives from several leading American technology companies have signed a voluntary artificial intelligence safety agreement aimed at creating common safeguards for the development and deployment of increasingly powerful AI systems.
Trump Unveils Voluntary AI Safety Accord
The agreement, signed at the White House on September 29, establishes a four-layer framework for AI companies to monitor and review safety risks.
The measures include internal safety controls during model development, dedicated teams responsible for checking those safeguards, independent external audits and board-level oversight of safety findings.
Trump described the agreement as “almost like a constitution” for the technology sector, while stressing that it is morally rather than legally binding.
Meta, Google and Nvidia Chiefs Back AI Pact
Executives participating in the initiative included leaders from major technology and AI companies such as Meta, Google, Nvidia, OpenAI, Anthropic and companies associated with Elon Musk.
Meta CEO Mark Zuckerberg described the agreement as an important industry-wide step designed to strengthen public confidence in advanced AI systems.
The pact comes as technology companies face growing scrutiny over cybersecurity risks, autonomous AI agents and the possibility of advanced systems behaving unpredictably during testing.
AI Companies to Rely on Self-Regulation
The agreement does not create new federal penalties or mandatory government inspections. Instead, participating companies will largely implement and evaluate the safeguards themselves, alongside independent auditors.
Trump has continued to favour a relatively light regulatory approach, arguing that excessive restrictions could slow American innovation and weaken the country’s technological position against China.
The administration has indicated that stronger rules could eventually follow as AI capabilities and associated risks continue to evolve.
