@imaxxs My experience, by harnessing a frontier model that is most likely going to keep track of the multi step plan, and keeping history of past successful execution as memories
To be clear, what the @DOGE team and @USTreasury have jointly agreed makes sense is the following:
- Require that all outgoing government payments have a payment categorization code, which is necessary in order to pass financial audits. This is frequently left blank, making audits almost impossible.
- All payments must also include a rationale for the payment in the comment field, which is currently left blank. Importantly, we are not yet applying ANY judgment to this rationale, but simply requiring that SOME attempt be made to explain the payment more than NOTHING!
- The DO-NOT-PAY list of entities known to be fraudulent or people who are dead or are probable fronts for terrorist organizations or do not match Congressional appropriations must actually be implemented and not ignored. Also, it can currently take up to a year to get on this list, which is far too long. This list should be updated at least weekly, if not daily.
The above super obvious and necessary changes are being implemented by existing, long-time career government employees, not anyone from @DOGE. It is ridiculous that these changes didn’t exist already!
Yesterday, I was told that there are currently over $100B/year of entitlements payments to individuals with no SSN or even a temporary ID number. If accurate, this is extremely suspicious.
When I asked if anyone at Treasury had a rough guess for what percentage of that number is unequivocal and obvious fraud, the consensus in the room was about half, so $50B/year or $1B/week!!
This is utterly insane and must be addressed immediately.
@g1mike@truthpole That looks like what I saw yesterday high in the sky over the San Francisco Bay Area, captured with an iPhone 15 Pro (3x). I wish I’d had a telephoto lens! It hovered above the clouds and was stationary for a while before moving south, where I eventually lost sight of it.
The more that is spent, the worse it gets, because the money doesn’t go to the homeless, it goes to the hundreds of “charities” that then treat the homeless as sources of revenue.
The more homeless there are, the more money these organizations get, so their incentive is to increase, not decrease, homelessness!
Whatever you incentivize will happen.
Spent the evening exploring o1-preview, and it struck me: in a world where AI is everywhere, it’s no longer about machines enhancing us—it’s about us enhancing the machines to truly stand out.
New Logcaliper update is live! Discover how a gram of DNA can store 9 million Blu-ray discs worth of data. Imagine 4,702 years of 4K YouTube videos in a tiny space or storing Wikipedia's content 10 million times. Dive into data like never before with our Compare Mode. #Logcaliper #DataScience #SIEM
https://t.co/tlWA83RBGT