I did look at those OIG reports because it sounded like you had a compelling argument and I wanted to learn more about it (I try not to assume every decision by current administration is bad). The amounts in the OIG reports range from around 30k-200k. The yearly budget for NSF 2024 was 9 billion. 200k/9b = 0.00002 or 0.002% - i.e. practically nothing. Additionally, each of those reports covered 3 years on average, so divide that by three. I can't even discern from the OIG reports if the findings were specific to NCAR. If they don't have anything to do with NCAR, then how are they relevant at all to this discussion?

Do you have more compelling data/evidence than the cited OIG reports which shows misappropriation of funds by NCAR?

This is probably not an accurate analogy. Imagine the first calculators that came out. They were probably correct 99.99% of the time (with only the more obscure functions possibly being buggy in the very earliest days of calculators). These LLMs we're dealing with often provide incorrect results. Imagine trying to do math on a device where it tells you that sin(239834+29348)=.497823 (Completely made up numbers all around). The answer looks correct (sin values are always between -1...1) but just looking at it, I'd have no idea if it's correct or not. Calculators you could trust. LLMs not at all.

All of these units will have to phone-home to a central server in order to get real-time updates on the status of the load on the grid. That data will be used to determine if it's ok to turn on or not. I can envision a plausible scenario where there is mass distribution/usage of these units, the grid starts relying on them (i.e. they don't plan for the extra load on the system any more), then some malware is deployed that exploits a vulnerability in these devices which essentially tells all of them simultaneously to turn themselves on which might result in rolling black-outs.

Worse - the malware could overwrite the firmware of these thermostats such that they are always on 100% for the opposite temperature you want (i.e. set it to "cool to 30 degrees in winter" and "heat to 120 degrees in summer") then semi-brick them so that no future firmware updates can be made. That would screw with the customers and the grid if it was deployed during a heatwave or an ice storm. Imagine trying to coordinate manually fixing hundreds of thousands of thermostats during a major weather event.

Oh you wanted a "no working fix"? Sure, that's still available - got plenty of those in-stock.

nt

Will you please share the name of the sleep clinic that helped you?

Ciphers -chacha20-poly1305@openssh.com

How well have humans survived all of those billions of years of change? Oh right... The earth will continue on just as it always has. Humans on the other hand...