Chat GPT AI
- Thread starter Sanrith Descartes
- Start date
-
Guest, it's time once again for the massively important and exciting FoH Asshat Tournament!
Go here and give us your nominations!Who's been the biggest Asshat in the last year? Give us your worst ones!
Sanrith Descartes
You have insufficient privileges to reply here.
- 44,495
- 120,674
We call this "tribal knowledge". My chief designer understood this well. Whenever I asked him about a part design he would just tap his head and say "Its all up here".
Sanrith Descartes
You have insufficient privileges to reply here.
- 44,495
- 120,674
Nah, its about politics. Notice with the exception of California, blue states audits are nearly non-existent. Imagine a state the size of NY being this low on the audit list.
- 1
- 1
Pasteton
Blackwing Lair Raider
- 2,733
- 1,919
Akshually if you look at Cali the highly densely populated blue areas aren’t audited as much either, it looks like the auditing is happening all in the red parts of Cali (by just area, cali is more red than blue I believe).
Source of this map? I have a feeling this is more to do with wealth and influence than politics. Going after some finance bigwig on the coasts means guaranteed lawyers and pushback which the irs doesn’t have the resources /is way too lazy to deal with. But that small business owner in a flyover city? Can make him roll right over
Source of this map? I have a feeling this is more to do with wealth and influence than politics. Going after some finance bigwig on the coasts means guaranteed lawyers and pushback which the irs doesn’t have the resources /is way too lazy to deal with. But that small business owner in a flyover city? Can make him roll right over
Sanrith Descartes
You have insufficient privileges to reply here.
- 44,495
- 120,674
Where in the U.S. Are You Most Likely to Be Audited by the IRS?
A new study shows dramatic regional differences in who gets audited. The hardest hit? Poor workers across the country.
projects.propublica.org
Where in the U.S. Are You Most Likely to Be Audited by the IRS?
A new study shows dramatic regional differences in who gets audited. The hardest hit? Poor workers across the country.projects.propublica.org
Rather than politics, this is reflective of auditing people receiving assistance in various forms. Seems like we should be happy about that. From your link:
"The five counties with the highest audit rates are all predominantly African American, rural counties in the Deep South. The audit rate is also very high in South Texas’ largely Hispanic counties and in counties with Native American reservations, such as in South Dakota. Primarily poor, white counties, such as those in eastern Kentucky in Appalachia, also have elevated audit rates."
Audits are lower in areas where people file straight 1040's and pay from withholding. No surprise.
Furry
🌭🍔🇺🇦✌️SLAVA UKRAINI!✌️🇺🇦🍔🌭
- 21,885
- 28,606
I just file and do the default deduction every year except one where I had a viral video that made me a ton of money. Deducted a bunch of electronics and travel expenses and paid 0$ in taxes on that money. Thought for sure it’d be my one audit, but nothing happened.
Sanrith Descartes
You have insufficient privileges to reply here.
- 44,495
- 120,674
There are a metric fuck ton of people receiving assistance in NY. Similarly their are a metric fuck ton of African Americans in NY.
Sanrith Descartes
You have insufficient privileges to reply here.
- 44,495
- 120,674
Sanrith Descartes
You have insufficient privileges to reply here.
- 44,495
- 120,674
Tuco
I got Tuco'd!
- 47,358
- 80,737
Tech doesn't really follow a hockey stick. It's more like two hockey sticks where the top one is inverted. Or more specifically, a sigmoid curve.
This curve naturally occurs due to physical limitations. Sometimes a new approach or material is discovered that allows the creation of an another sigmoid curve to reach higher.
In the case of many AI topics, this is because there are limitations to certain AI approaches that start to logarithmically slow down improvements before a critical threshold of usability can be reached. It's confusing to see a massive exponential growth followed by a slowing of improvement, but that's just the nature of it.
You can observe that with object detection from images. Where depending on the image database, getting to 80-90% accuracy happened quickly. Getting to 99% accuracy may need an entirely different technology other than the type of convolution networks.
Object Detection and Image Segmentation with Deep Learning on Earth Observation Data: A Review-Part I: Evolution and Recent Trends
Deep learning (DL) has great influence on large parts of science and increasingly established itself as an adaptive method for new challenges in the field of Earth observation (EO). Nevertheless, the entry barriers for EO researchers are high due to the dense and rapidly developing field mainly...
www.mdpi.com
- 5
Tuco
I got Tuco'd!
- 47,358
- 80,737
Another example of a sigmoid curve is battery density. This is more tractable because there are chemical limitations, where new chemical reactions may be developed to replace Li-on or someone makes an arc reactor or something, but otherwise there's only so much you can do.
- 2
Palum
what Suineg set it to
- 26,510
- 41,249
Tech doesn't really follow a hockey stick. It's more like two hockey sticks where the top one is inverted. Or more specifically, a sigmoid curve.
This curve naturally occurs due to physical limitations. Sometimes a new approach or material is discovered that allows the creation of an another sigmoid curve to reach higher.
In the case of many AI topics, this is because there are limitations to certain AI approaches that start to logarithmically slow down improvements before a critical threshold of usability can be reached. It's confusing to see a massive exponential growth followed by a slowing of improvement, but that's just the nature of it.
You can observe that with object detection from images. Where depending on the image database, getting to 80-90% accuracy happened quickly. Getting to 99% accuracy may need an entirely different technology other than the type of convolution networks.
View attachment 477297
Object Detection and Image Segmentation with Deep Learning on Earth Observation Data: A Review-Part I: Evolution and Recent Trends
Deep learning (DL) has great influence on large parts of science and increasingly established itself as an adaptive method for new challenges in the field of Earth observation (EO). Nevertheless, the entry barriers for EO researchers are high due to the dense and rapidly developing field mainly...
View attachment 477298
Yes I was discussing this issue the other day and also noted that holistically, the trend is to find utility for surplus capacity which further cannibalizes the prospect for advancements. Which is to say scope creep is real and endemic, diluting anything as soon as it reaches a point where it has economic utility.
E.g. Open AI spends significant resources filtering data and training its models to not be racist, which is not to the benefit of the core model development rather than religious adherence.
- 1
Mist
REEEEeyore
- 31,197
- 23,355
You don't understand the architecture of these models.
The layers that screen out things like racism are not the core model. Core model training is just a statistical model of word order. The core model is only built once, during the training run, and then it is one giant terabyte+ long number, until a new model is built in the next training run with a new, larger base training set.
The alignment layers are much higher up the stack than the core model, applying human feedback training. These alignment layers are constantly being adjusted through use, but the core model remains fixed. You could throw out, or replace, that layer, and apply a different layer in its place, and it would change the output but it would not change the core model.
- 1
Palum
what Suineg set it to
- 26,510
- 41,249
TheNozz
Karazhan Raider
- 7,894
- 40,199
I’ve theorized for several years now that Antichrist would pose as an AI.
I mean it wouldn’t be real AI but probably demonic interference through the AI.
- 2
- 1
