On October 30th, President Biden issued an Executive Order on “Secure, and Trustworthy Artificial Intelligence”. Here are some quick thoughts on what that means.
First, “Executive Orders” as you may recall, are declarations by the President directing Federal officials or administrative agencies to engage in, or refrain from, courses of action. In this case, President Biden directed several agencies to investigate and act upon different aspects of AI.
For example, this Executive Order asks the National Institute of Standards and Technology to set “rigorous standards” to ensure that AI systems are “safe, secure and trustworthy”. It also says that the Department of Homeland Security will “apply those standards to critical infrastructure sectors and establish the AI Safety and Security Board”. It says that the Department of Commerce “will develop guidance for content authentication and watermarking to clearly label AI-generated content.” And it directs the National Security Council and the White House Chief of Staff to develop “a National Security Memorandum that directs further actions on AI and security.”
In many respects, the document starts the wheels rolling on investigating how we can protect ourselves from AI while at the same time investigating how we can use it to make our lives better. And it wants to start those investigations of AI in the areas of national security, biological materials, fraud, cybersecurity, privacy, civil rights, consumer protection, patient rights, students, workers, government, and international leadership.
If that sounds like a lot to you, well that’s comforting because it sure sounded like a lot to me.
And how will it do these things? And to what end? Well it doesn’t say, which yeah, I get it – this is the beginning and if you could lay out every nut and bolt you probably wouldn’t need an Executive Order in the first place.
But because it doesn’t have any specifics and because it looks like it’s trying to do everything and the kitchen sink, it’s hard to not to look at this as a very political document. Especially when you remember that the UK AI Safety Summit is taking place on November 1 and 2 at Bletchley Park (which you may remember as where Alan Turing spent much of World War II breaking codes and inventing computers).
So yeah, sure, how could even the most non-political document in the world not look like a political document in these times where this topic is concerned. Especially one that is trying to touch as many different parts of our society and culture as this one is because, well, that’s what AI will do too…
That said, one would like to think that the Executive Order’s diversity of focus and purpose is actually its strength; that if you gather people who specialize in lots of different areas all studying the impact of the same thing on their particular area of expertise, you’ll get a robust and innovative series of solutions when they all come back together again.
Does government work like that? Probably not but hey there’s always a first time.
Additional Reading:
The Executive Order on Safe, Secure and Trustworthy Artificial Intelligence
Forbes: What Biden’s New Executive Order Could Mean For The Future Of AI
New Atlanticist: Experts react: What does Biden’s new executive order mean for the future of AI?