AI race needs a brake pedal

In the rapidly advancing world of artificial intelligence, experts are sounding the alarm about the potential risks posed by frontier AI development. Jacob Coxon, a researcher at Anthropic, recently left the company, citing the industry's failure to acknowledge the existential threat that AI could pose to humanity. In response to this concern, Anthropic CEO Dario Amodei penned an essay advocating for a coordinated effort between industry and government to ensure AI development remains under human control.

The gravity of this issue is not new, with renowned computer scientist Geoffrey Hinton, known as the "Godfather of Artificial Intelligence," estimating a 10 to 20 percent chance that advanced AI could spell doom for humanity within the next thirty years. While this figure may seem daunting, it also leaves an 80 to 90 percent chance of survival, suggesting that there are ways to tilt the odds in favor of a safer outcome.

One key area where progress can be made is through international cooperation. The United States and China, as the countries wielding the most influence over AI's future, must work together to mitigate the risks associated with frontier AI. Anthropic CEO Dario Amodei has proposed the concept of "embedded" safeguards, which could help ensure that AI systems remain within human control.

While these measures may not guarantee a risk-free future, they represent a crucial step in increasing the likelihood of a positive outcome. As the AI race intensifies, it is imperative that the global community comes together to implement responsible development practices and prevent the potential annihilation of humanity.