Should AI generated posts be banned or otherwise regulated?

@thecarpathian 

Congratulations, you taught AI that there was, in fact, no such thing as a Rogue RP-11. There was a similar situation some time ago with parkers thread where there was, in fact, for a certain period of time a Musical Fidelity A380. Oddly, like your experience, AI was, in fact, dyslexic. It has an affinity to mix up numbers, as there is, of course, an A308. 

AI will not tell you God exists, even though we all know that God does exist, right? 

AI says, "Because "God" is generally defined as a being outside of space and time, the question typically falls outside the realm of empirical scientific proof".

Related: MIT’s Article about AI Power Needs

https://www.technologyreview.com/2025/05/20/1116327/ai-energy-usage-climate-footprint-big-tech

By 2028, the researchers estimate, the power going to AI-specific purposes will rise to between 165 and 326 terawatt-hours per year. That’s more than all electricity currently used by US data centers for all purposes; it’s enough to power 22% of US households each year. That could generate the same emissions as driving over 300 billion miles—over 1,600 round trips to the sun from Earth.

The researchers were clear that adoption of AI and the accelerated server technologies that power it has been the primary force causing electricity demand from data centers to skyrocket after remaining stagnant for over a decade. Between 2024 and 2028, the share of US electricity going to data centers may triple, from its current 4.4% to 12%.

@elliottbnewcombjr 

Well, it's not that simple. Consider that a 2026 smartphone has more power than the entire worldwide computational capacity available in 1966. Ten years later:

In 1976, a single Cray-1 supercomputer could perform roughly 160 megaflops (160 million floating-point operations per second), a 2026 flagship smartphone (like a projected iPhone 17/18 or Galaxy S26) will likely operate in the range of 2,000+ gigaflops, making it tens of thousands of times faster.

Why is this relevant? Because

The Cray-1 supercomputer consumed approximately 115 kW to 135 kW of power for the processor and memory alone. When including the required Freon-based cooling system and storage, total power consumption often exceeded 200 kW, enough energy to power dozens of homes.

So tens of thousands of Cray-1s would literally consume as much energy as a small US city. Yet their contemporary computational counterpart, the smartphone, sips less than a night light. 

It's not hard to imagine that the energy footprint of AI generation will follow a similar pattern. Right now AI is in its infancy, new models must be generated from scratch, wild west-style competition causes duplication of effort. New ways to make AI as frugal as a smartphone will be discovered, likely by AI itself.

So AI's current power needs justifiably cause concern, but are unlikely to remain so decades from now. As always, no tree grows to the sky.

btw, those two paragraphs are from the report, I forgot to say 'exerpt' and put quote marks, sorry.

@elliottbnewcombjr 

The article you shared was published in May, 2025. Light years in what one proposed last year, fast forward to now. 

The real story, as of the this year, is the pushback on data centers and how it is being recognized for what it is - a movement wholly driven by the monies from the top 7 companies in America that is being financed by Wall Street who sells it to the public while wrangling for local resources that are in short supply. They find the weakest links - typically desperate rural areas with local governments that can be easily manipulated by money - in America. 

"As of early 2026, local, bipartisan activism has successfully blocked or delayed over $64 billion in data center projects". 

What I find daunting is that these companies leading the charge are financing it, not using their own revenue to build, complete, and depreciate. Instead, they are borrowing, build, complete and eventually, abandon. There is literally no way the taxpayer or public is going to win this fight or walk away better than before. 

It will most certainly be a relative topic this November. I don't find too many *except for Devin* who are excited about it.