TECH GIANTS SIGN AI SAFETY PACT AT WHITE HOUSE

Executives from Meta, OpenAI, Anthropic, Google and Nvidia have agreed to a voluntary framework of artificial intelligence safety standards following a meeting with US President Donald Trump at the White House on Tuesday. The accord marks a significant moment in Washington’s efforts to shape the governance of rapidly advancing AI technologies.

The agreement commits participating companies to strengthening internal controls and collaborating with independent auditors to ensure their systems function as intended. This includes measures designed to prevent AI tools from being misused to hack or infiltrate other technical systems. By pledging to adopt these safeguards, the firms are signalling a willingness to balance innovation with responsibility, amid growing public and political scrutiny of AI’s potential risks.

The White House meeting reflects mounting concern over the unintended consequences of AI deployment, from cybersecurity vulnerabilities to ethical dilemmas in automated decision-making. President Trump emphasised the importance of industry-led initiatives that can complement government oversight, framing the pact as a proactive step to protect national interests while sustaining technological leadership. The administration has been under pressure to respond to calls from lawmakers, academics and civil society groups for clearer guardrails around AI development.

For the companies involved, the voluntary standards are both a gesture of accountability and a strategic move to shape the regulatory conversation. By agreeing to independent audits, they are opening themselves to external scrutiny, which could help build public trust in their systems. At the same time, the pact allows them to retain flexibility in how safeguards are implemented, avoiding the rigidity of formal regulation. Industry observers note that this balance may prove crucial in maintaining competitiveness while addressing legitimate safety concerns.

The inclusion of Nvidia, a hardware powerhouse whose chips underpin much of the AI revolution, alongside software leaders such as Google and OpenAI, underscores the breadth of the commitment. It suggests recognition that safety must be embedded across the ecosystem, from infrastructure to applications. Anthropic and Meta’s participation further highlights the diversity of approaches within the sector, united by a shared interest in mitigating risks.

While the agreement is voluntary, its symbolism is significant. It represents a rare moment of consensus among rival firms, and a willingness to engage constructively with government. Whether these standards will prove sufficient in practice remains to be seen, but the pact sets a precedent for collaboration at a time when AI’s trajectory is both promising and perilous. The White House has signalled that it will continue to monitor progress, leaving open the possibility of more formal regulation if voluntary measures fall short.

Meta, OpenAI, Anthropic, Google and Nvidia have agreed to voluntary AI safety standards after meeting President Trump at the White House. The pact calls for stronger internal controls, independent audits and safeguards against misuse, marking a significant step in balancing innovation with responsibility amid growing scrutiny of AI risks.