Is the AI safety debate about safety or control?
Not everyone agrees with Amodei's call for globally coordinated action for AI safety.
The ongoing AI safety debate has sparked a discussion about whether the focus should be on safety or control. High-level executives from different AI companies have weighed in on the matter, with some advocating for a deceleration of AI advancements to ensure adequate guardrails are in place. OpenAI CEO Dario Amodei penned a nearly 4,000-word essay urging for a slowdown in AI development, suggesting an international collaboration between companies and governments for safe deployment.
However, rival AI executives like OpenAI's Sam Altman and xAI's Elon Musk have endorsed Amodei's plan. Meta CEO Mark Zuckerberg, too, pushed back the release of his AI model due to safety concerns and has expressed hope that companies would figure out security measures on their own. Reddit co-founder Alexis Ohanian also expressed optimism about companies addressing AI risks independently.
Despite calls for public-private collaboration, some voices seem content with letting businesses set their own rules, as evidenced by the creation of an AI standards organization by major AI companies. Critics argue that this move could lead to "regulatory capture," where powerful companies manipulate regulations to favor themselves and disadvantage smaller competitors.
This concern extends to China, where Chinese authorities view American AI labs as seeking to disrupt global AI governance.
Written by urgent.news from TechCrunch's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
