Will Musk’s use of AI to target government spending cuts become a Westworld moment . . . ?

Photo above - Elon Musk explaining his plan to use AI to make government more efficient . . .

This could be the “make or break” moment for AI. Using a few lines of cleverly written code to ferret out everything that's wrong, both in private sector and government workplaces. It doesn't matter whether those SNAFUS are accidental on purpose. Musk could either be a hero, or end up souring the nation on AI, for years to come. (see link at bottom)

A true story: I have a friend who works the mortgage department of a “too big to fail” bank". She uses ChatGPT (increasingly) to handle her workload. To complete federal, state and internal bank forms. To compose written communication with customers. To write her emails, and report to her boss on the status of pending mortgages.

My friend fails to see the threat to her job.

She is indifferent about the potential that someone else (at a much lower salary) could do the work of 5 mortgage bankers through ChatGPT. Or with a competing AI system: Gemini, DeepSeek, or Rehoboam (from WestWorld). Although there could be pushback by some people against using a system named after the King of Judea. In any case, the mortgage process will be vastly simplified.

“Gemini . . . I’d like to apply for a $350,000 30-year fixed rate loan on the condo for sale at 742 Evergreen terrace . . .”

(AI assistant*). “Working. Gathering 5 years of tax forms, proof of citizenship, work and school attendance records. Negotiating with bank. Good news, your loan has been approved. Your rate is 7.5%. Closing will take place tomorrow at 10am. Would you like me to order a croissant and cappuccino for you to pick up on the way?”*

Musk’s use of AI will be startlingly similar. “Give me a list of all GS 10 or lower federal workers who have absences without a doctor's note. Tardiness. Refused to return to the office. Unpaid taxes, child support, or alimony. Pending charges for anything other than a traffic violation. Who is in the bottom 25% of quarterly performance reviews. Now rank these from worst to best, according to enterprise value metrics.”

(ChatGPT) “Working. I found 358,732 workers who meet these parameters. Would you like me to print this for you?”

(Musk) “No . . . don’t bother with all that. Just send them buyout emails.”

It's funny because it’s true.

Of course there will be lawsuits. Court challenges, perhaps filed with the assistance of AI. The natural response will be to employ an AI judicial system to dismiss all filings which use an incorrect form, are in the wrong jurisdiction, or fail to cite relevant statutes or case law as precedent.

“Sorry. Your civil suit was dismissed. Would you like me to order a croissant and cappuccino for you from the courthouse coffee shop on your way out?”

It won’t stop with mortgages and quarterly personnel performance assessments, though. AI could quickly examine the nation’s worst preforming school systems and fire those administrators. Then provide all classrooms with updated curriculums. Teachers who resist could be replaced with online instruction.

But here’s what will be my favorite use of AI: to prevent Covid 19 from ever happening again.

I don’t mean stopping the next virus escape from a Chinese military lab, or which spontaneously appears in chickens, cows or pigs, or cave bats. AI cannot prevent genetic mutation. I mean the administration of Covid relief/stimulus funds. Imagine how much better things would have been if we hadn’t been conned out of $400 billion in grants by fraudsters, rappers and influencers. People who used that money to buy Lamborghini’s and Vail ski condos. (see 2nd link at bottom). Possibly with the collusion of government grants administrators . . .

I’m just sayin’ . . .

Elon Musk’s DOGE is feeding sensitive federal data into AI to target cuts

Orange County man who bought luxury cars with COVID relief funds sentenced to prison - Los Angeles Times

AuRevoir · 36-40, M

People act as if AI is something new and that it hasn’t been being used to do “things” since their primary use in video games. Which has been around forever.

People use things as basic as spreadsheets as tools to do things, and essentially that’s what you can program top tier big money AI to do. Have it run a diagnostic, spreadsheet algorithm, give it the principal commands and discrepancies to look for. Have it perform quadruple checks on its findings. Then the human eyes will look it over, and confirm the biggest discrepancies first.

It’s not that difficult to implement.

And after human eyes look over the ones and confirm what they wish to remove. You then remove it from the list, and have it look all over the numbers again. Basically the AI will be running simple programming with the next level efficiency of what current modern technology has developed with advance AI’s which I’m sure Musk’s own Tesla technology company has one of the most advanced ones in the world.

He’s already used it before to get rid of all the spam bots that former Twitter was infested with. And proved to all the major companies that had spent their money over the years advertising on Twitter, that the people in charge were aware of the bots and even responsible for the bots. But allowed them anyway. Thus cheating and deceiving advertisers out of their money with their con scheme. To ironically no outrage. And instead only crickets. You’d think there would have been a lot more lawsuits after that kind of corruption scandal was exposed. But surprisingly there was no yelling at the former owners. Or if there was, my news feed never caught wind of it…

Basically it will be an AI that’s dedicated and designed around accomplishing only certain types of things. It won’t be like how the populace uses something like chatgpt for “everything” which will make the refined abilities of the AI even more advanced at accomplishing its job.