Today, President Biden signed an order to regulate the artificial intelligence (AI) business. The order says that AI must promote “civil rights” and “equity,” which is the left’s concept for diversity, equity, and inclusion (DEI).
The White House says it will send advice to landlords, government contractors, federal welfare programs, and other groups on how to use AI in a way that is less likely to be biased against people of color or other groups. Furthermore, enforcement measures will be taken. According to the White House fact sheet on the executive order, “training, technical help, as well as coordination involving the Dept. of Justice and the Federal civil rights offices on the most effective methods for the investigation and prosecution of civil rights offenses related to AI” will be made available.
A major focus will also be put on utilizing AI in the criminal justice system, with plans to create “the most effective practices on the use of AI in risk assessments, surveillance, crime forecasting and predictive policing, parole and probation, pretrial release and detention, and forensic analysis.”
There is an unsaid fear that letting AI fully look at certain kinds of data, like crime records and credit scores, could lead to unfair results for some groups. This worry is at the heart of the academic field of machine learning fairness, which is a mix of critical race theory together with computer science. Breitbart News has written a lot about this topic before, and this portion of the executive order appears to build on it. In the end, companies that use AI may sometimes have to stop them from coming to too exact conclusions so that they don’t break U.S. civil rights laws, especially the concept of disparate effect.
There are a lot of other important things in the executive order as well. AI companies will have to use watermarks to ensure that people can tell the difference between content made by AI (especially photos and movies) and content not made by AI.
It also gets the government involved in testing AI for safety. The Dept. of Homeland Security (DHS) will set up an “AI Safety and Security Board” to make sure that key infrastructure sectors comply with those safety requirements. The National Institute of Technology and Standards will come up with rules for testing the safety of newly developed AI systems.
Given DHS’s main role in censoring and targeting Americans, the fact that it is involved in regulating AI may worry people who fight for online freedom. In the past, DHS sub-agencies have used cybersecurity as an excuse to block “disinformation” and make plans for “deradicalizing” Americans who are conservative.
The Departments of Energy as well as Homeland Security are also going to “deal with cybersecurity risks, chemical, biological, radiological, and nuclear threats, as well as AI systems’ threats to critical infrastructure” in terms of security. The White House brief also says that agencies that fund projects in the life sciences will set standards for biological synthesis projects. This is to make sure that AI doesn’t make it easier to make biological materials that are harmful.
Comments are closed.