Introduction - AI Presentation 2026

1. What the Presentation Covers

Artifical Inteligence models have be developed to play grand master level chess and go. Others have learned to create protein folding and weather prediction models that are as good as, or better, than existing methods.

In this presentation we will be looking at a subset of AI models called Large Language Models (LLMs). they:

answer questions about a large corpus of knowledge that may include much of the web, portals holding scientific papers and code, forum sites and Wikipedia
help building slide decks, pdf documents, and web pages
Carry out tasks like testing code and sorting email

Specifically we cover:

how people like you and businesses like ClarkLinsey use LLM AIs
What models cost and benefits and deficits that are inherent to LLM AIs

The presentation will survey the surface of these but cannot burrow too deeply into details.

#	Page	Focus
1	Introduction	This page: what the presentation covers and where its figures come from
2	Use Examples	What people and companies do with the models today
3	How chat bots work	What happens between your question and the reply
4	LLM AI Command Line Interface (CLI)	Working with an LLM from a terminal, with access to local files and tools
5	LLM AI Agents	Working with an LLM from a terminal, with access to local files and tools
6	LLM AI Benefits	What the models return, to one person and to a population
7	LLM AI Deficits	How the models fail, who absorbs the damage, and who answers for it
8	Data Centers	The buildings the models run in, where they sit, and what the machines actually do
9	Economics	What the companies earn, what they spend, and who supplies the difference
10	How does AI work	The mechanism behind the predictions, told without mathematics

2. Source Materials

Eight documents sit in this folder alongside the page. The two financial files and the three data center files supply the figures the pages quote. The three topic papers carry the arguments.

Document	Contents
`LLM_Business_Uses.docx`	How companies use the models, and where the money lands
`Benefits of LLM AIs`	Effects at the global and societal level
`Global_Liabilities_of_LLM_AIs.docx`	Harms, costs borne by outsiders, and questions of accountability
`Anthropic_Financials_2021-2025.pdf`	Revenue, investment, and spending, 2021 through 2025
`OpenAI_Financials_2015-2025.pdf`	Revenue, investment, and spending, 2015 through 2025
`DataCenters_8.3_Capacity_Evolution.pdf`	How much computing capacity exists, and how fast it grew
`DataCenters_8.4_Geographic_Distribution.pdf`	Where the buildings stand, and what put them there
`DataCenters_8.5_Workload_Distribution.pdf`	How the machines divide their time between training and inference

3. Reference Articles

This presentation shows no code. A separate set of tutorials, written for programmers, covers how a program sends a question to one of these models and handles the reply. The Tutorials link above opens that index.