AI NEEDS STRONG REGULATION

SUMMARY: Artificial intelligence (AI) needs strong regulation because its effects are far-reaching and powerful. U.S. Representative Khanna states that a robust federal agency is needed to regulate AI and has put forth seven principles for a new social compact to address the issues of AI and inequality.

(Note: If you find a post too long to read, please just skim the bolded portions. Thanks for reading my blog!)

(Note: Please follow me and get notices of my blog posts on Bluesky at: @jalippitt.bsky.social. Thanks!)

My previous post provided an overview of the need for strong regulation of Big Tech companies and focused on artificial intelligence (AI). It also shared an innovative proposal from U.S. Senator Sanders to regulate the AI industry through public ownership of 50% of the big AI companies. This post will go into more depth on the need to regulate AI and principles for ensuring that AI produces public benefits rather than solely private wealth.

AI needs strong regulation because its effects are far-reaching and powerful. AI is having dramatic impacts on the news we get, the entertainment we enjoy, and the social media we consume. It’s affecting our jobs, sometimes to the point of getting laid off and replaced by AI. It’s having negative effects on education and mental health. Its data centers are affecting local communities and increasing costs for electricity and water, as well as having serious negative environmental impacts. The paper-wealth AI is generating is skewing our society and politics, while also creating a financial bubble that could burst with serious negative effects on financial markets and our economy. Some of AI’s worst aspects may be in its use by the military and its ability to generate pornographic images.

A key ingredient of AI’s implementation has been the combination of huge databases (including our personal information) and massive computing power. In many cases, it isn’t the algorithms and their “intelligence” that are new, but simply AI’s use of them with huge databases and computing power. In many ways, this is a brute force approach that is reflected in the industry’s need for huge data centers using massive amounts of electrical power to operate and of water for cooling.

The result is an amazing ability to manipulate us, for example, our shopping and our addiction to social media. It also gives the AI and related companies or governments, incredible power to surveil us – to know where we are and what we’re doing. Make no mistake, we and our data are the product, and we are very much for sale.

If AI were being implemented with the best of intentions, as a beneficial innovation, the amount of misinformation would be declining and not growing; faked images, text, and speech would be detected and not disseminated; and inflammatory rhetoric and hateful content would be identified and blocked. AI certainly has the capability to do these things if they were the goals of its creators and implementers. However, the industry’s goals are clearly to maximize profits and wealth. Many experts who are deeply involved with and knowledgeable about AI are sounding warnings about possible catastrophic results. [1]

To counter the growing public unease with AI, the AI companies and their incredibly wealthy founders and investors are spending tens and sometimes hundreds of millions of dollars on election campaigns and lobbying. As a result, there often is support for the industry among policymakers. The Trump regime is promoting the U.S.’s AI world leadership as a national security priority. Trump issued an executive order directing his Attorney General to sue states that try to regulate AI. His key policy makers for AI have substantial conflicts of interest through investments in AI companies. [2] There have been significant government investments in AI for its initial development and its deployment. There are government subsidies at the federal, state, and local levels for the construction of AI’s data centers, for the chips needed for AI computing, and even for the exports of AI companies. There’s a $1 billion federal government loan for the restarting of the Three Mile Island nuclear power plant to provide electricity for a data center. This is oligarchic crony capitalism at work.

U.S. Representative Ro Khanna (D-CA) states that a robust federal agency is needed to regulate AI and has put forth seven principles for a new social compact to address the issues of AI and inequality, which AI will exacerbate if not well regulated. Here are his principles: [3]

• Ensure AI augments human capabilities and doesn’t just eliminate jobs.
• Require that workers benefit from AI-driven productivity gains through higher wages, profit sharing, and a shorter work week. Americans should also get paid for the data they generate that AI uses.
• Fix our tax system that currently subsidizes robots with accelerated depreciation and disadvantages human workers via payroll taxes.
• Provide AI-displaced workers jobs in public service, as FDR did during the Great Depression.
• Require data centers to serve the communities that host them, from supporting schools and libraries, to supporting local businesses, to avoiding increases in electricity and water costs, as well as negative environmental impacts.
• Require social media platforms to stop AI algorithms from spreading hate and other harmful content. Require them to support interoperability so users are not trapped by monopolistic control but can switch and connect across platforms.
• Regulate AI with a robust federal agency so it’s used to enhance humanity and not harm it.

For lots of good news see Jess Craven’s Chop Wood Carry Water blog’s most recent good news Sunday post here.

---

[1]      Khanna, R., 4/8/26, “AI for the people,” The Nation (https://www.thenation.com/article/society/ro-khanna-ai-democracy-blueprint/)

[2]      Glickman, S., Kak, A., & Myers West, S., 4/8/26, “The great AI grift,” The Nation (https://www.thenation.com/article/economy/ai-crony-capitalism-grift/)

[3]      Khanna, R., 4/8/26, see above