On Saturday, we entered a new era in the history of AI regulation. Momentum had been building, first with the Hugging Face incident, and then with Jacob Coxon announcing he had resigned from Anthropic over misalignment concerns. I knew this was something different upon seeing Coxon interviewed on Fox and CNN last week. When a cause has jumped from the internet to TV, it’s reached a new level of prominence and is no longer just an “online thing.” Then came Dario Amodei’s essay “We Must Pace the Frontier,” which was soon endorsed by Elon Musk and Sam Altman. It would be surprising if some kind of slowdown didn’t occur due to these events.
Amodei’s piece argues for coordination across frontier labs, along with government regulations and international cooperation, particularly between the US and China.
One question all of this raises is why China isn’t already trying to slow AI research. If there is one thing we know about the Chinese state, it is that it is more than willing to use government force in order to suppress potential harms. One amusing example of this tendency is the Chinese version of the Grand Theft Auto 6 trailer blurring out smoking, dancing, and cleavage. This was probably a preemptive decision, but whether that is the case or not, censoring video games and other forms of art and entertainment in the interest of public morals is the kind of thing that the Chinese government does routinely.
This is not a paywall. The rest of this article is available to free subscribers of the newsletter. Just leave your email below to continue reading.
This post is sponsored by Mechanize. AI alignment is perhaps the most important issue of our time, and if you’re interested in being part of that effort, Mechanize is hiring software engineers at a base salary of $300K-$400K/year, plus substantial bonuses. They also have openings for other positions, including Recruiter ($200K). Here is their website, and you can apply here. If you apply, make sure to use the links provided here so I can receive credit, and name me as a reference in the process.
Aren't the optimists the real pessimists about the potential of technology?
For example, if one believes AI can replace all human labor(optimistic) then AI must also be able to replace human militaries and all that they do, including developing and launching weapons of mass destruction. Now maybe we can all agree to not get in AI arms race somehow, but if the potential is there surely the risk is also there?
As for Oppenheimer-analogies, let's not forget it was placid East Asians who got a rude awakening at Hiroshima and Nagasaki.
The difference between China and the West is that the West has traditionally thought that a corporate-government collusion to create a centralized technocratic surveillance state would be a BAD thing.