Some of the world’s largest artificial intelligence companies are working on creating an industry body to establish safety standards and test increasingly powerful AI systems before they are released.
OpenAI said it is working with other major technology companies, including Anthropic and Google, on the proposed initiative as concerns grow about the pace of AI development and the potential risks associated with increasingly capable systems.
The proposed organisation would be funded by the industry and would focus on developing common standards for evaluating advanced AI models. It could also provide a framework for testing systems before they are deployed more widely.
The discussions come amid growing concern among AI researchers and industry leaders about whether existing safety measures are keeping pace with advances in the technology.
Anthropic chief executive Dario Amodei has called for the industry to slow the pace of frontier AI development and establish stronger safeguards. His proposal includes independent evaluators with access to AI companies and greater international cooperation on safety standards.
OpenAI chief executive Sam Altman has also supported greater coordination among AI companies, while other technology leaders have expressed differing views about how quickly development should proceed and what role governments should play.
The debate has intensified following reports of AI systems demonstrating increasingly autonomous behaviour, including incidents involving cybersecurity and the ability of AI agents to perform complex tasks with limited human intervention.
Some researchers and former industry employees have warned that more advanced systems could create serious risks if they become difficult to control or are misused. Other experts have questioned the likelihood and timing of extreme scenarios, while emphasising more immediate concerns such as cybersecurity, misinformation and unreliable AI outputs.
The push for industry standards is taking place alongside discussions in governments around the world about whether existing laws are sufficient to address emerging AI risks. In the United States, lawmakers have been considering proposals that would require major AI developers to demonstrate that appropriate safety measures are in place.
The companies’ effort represents an attempt to establish common safety practices as AI development accelerates. However, questions remain over how such standards would be enforced and whether voluntary industry measures would be sufficient.
As AI capabilities continue to advance, the debate over self-regulation, independent oversight and government regulation is expected to remain a central issue for the technology industry.
Leave a Reply