File size: 2,161 Bytes
1d5ebc8
 
9c0f971
69d9c2b
 
 
 
1d5ebc8
9c0f971
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7f54c5b
 
 
 
df1eaf6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
---
license: cc-by-4.0
pretty_name: Finance and AI Statistics, Source-Backed
tags:
- finance
- ai
- statistics
---

# Finance and AI Statistics (Source-Backed)

A small, curated dataset of statistics on AI adoption in finance, AI financial model risk, and spreadsheet reliability. Every figure links to a primary or authoritative source: an academic study, an official report, or reputable press for an event. No figure without a traceable source.

## Why this exists

Finance and AI statistics get recycled across blog posts until the original source is lost, and often until the number itself has drifted from what the study actually said. This dataset keeps the figure, its exact scope, and its primary source together, so it can be cited without laundering a number through five secondary blogs.

## Columns

metric: what is measured. value: the figure, as the source states it. statement: one-line context, with nuance preserved. scope: population, sample, period. year: reference year of the figure. source_org: the primary source. source_url: link to the primary source.

## Scope and inclusion

A row is included only if the source is primary or authoritative, it is linked, the figure is quoted as stated with no sense-changing rounding and no merged studies, scope is given, and nuance is preserved. A contributing factor is not written as a cause.

## Maintenance

Reviewed every 6 to 12 months. Dead links are re-sourced or removed. This mirrors the fuller, annotated library maintained by Layerz, at https://app.layerz.cc/resources : AI in finance, AI financial model risk, spreadsheet errors and risks, and fractional CFO statistics.

## Attribution

Compiled by Layerz, https://app.layerz.cc . CC BY 4.0: reuse with attribution.

## Related

- [layerz/finance-engineer-postings](https://huggingface.co/datasets/layerz/finance-engineer-postings): first-party corpus of every job posting we could verify for an engineering role inside a company's own finance function. 37 requisitions across 26 employers, coded by hand, with the sampling frame and adjudication rules published. Analysis: https://app.layerz.cc/resources/finance-engineer-statistics