Trump’s AI push: Tech giants agree to voluntary safety standards amid rapid AI development

US President Donald Trump met with several tech executives at the White House on September 29 to discuss rapid AI development and growing concerns. After the meeting, Trump announced that companies had agreed to follow voluntary AI safety standards. However, the agreement comes with few clauses that tech companies need to follow.

As per the AI accord, tech companies must have independent auditors check AI systems to see whether they are behaving as their developers intended. The companies have also agreed to ensure their AI system do not accidentally hack technical systems they were not supposed to access.

Must read: ‘Sundar is a monster, nobody knows him’: Donald Trump has a backhanded compliment for Pichai

Trump said the agreement would help protect the public and strengthen collaboration on AI safety. However, the voluntary framework does not give the government legally binding oversight powers.

“It's almost like a constitution, in a way,” Trump said about the agreement. “And the biggest people in the world signed that, and I signed it as president. And it really is a form of protection,” he added.

AI safety voluntary framework: What it mentions

Trump and tech leaders, including Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman, SpaceX CEO Elon Musk, Nvidia CEO Jensen Huang, and others, have signed a new AI safety agreement titled Joint Commitment on Frontier Responsibilities. The agreement highlights a set of AI safety rules that tech companies must follow to ensure public safety.

The agreement, titled the Joint Commitment on Frontier Responsibilities, outlines a set of measures for companies to monitor their AI systems and assess potential risks.

Under the framework, companies have agreed to strengthen internal safety monitoring during AI training, establish teams to review those controls, and allow external auditors to evaluate their safety measures.

Must read: 'AI is like fake news': Trump renames artificial intelligence as 'super intelligence'; here's why

The document requires companies to form an internal board that will review safety controls during AI training. However, companies are not obligated to publish the findings of evaluations.

Furthermore, companies must also bring in external auditors to evaluate safety measures to ensure AI systems operate as aligned. However, the agreement is non-binding and does not establish government oversight, which means companies are allowed to select their own evaluators, decide whether to publish findings, and no penalties are specified for non-compliance.

Trump backs data centre buildout

Apart from signing an agreement with AI companies, Trump also showed his continued support for data centre buildout. Over the months, data centres have faced backlash due to environmental concerns such as electricity consumption, power grids, water, and others.

Trump highlighted that tech companies will work with communities to address concerns surrounding data centre projects and provide support to areas where they are built. However, the issue has become a challenge for the administration ahead of the November 3 midterm elections.

In addition, Trump also made several changes to US government services and officially replaced the term “artificial intelligence” with “Super Intelligence.”