The Mining of training data
Training data is the information gathered across the internet used to create answers and images. Training data can take the form of Reddit, Wikipedia, personal social media pages, both free and paid media outlets — AI uses everything, to the tune of billions of pages each month for a given generative AI model.
How Large language models work
-
Model developers “acquire” content that is used as the primary input to train the model — acquisition happens through web scraping, shadow databases (including copyrighted content) and content uploaded by end users to various cloud-based platforms.
Copyrighted content
Web pages/content
Social media
Personal ID information
Confidential data
Photos and video content
-
Content is reviewed using automated and manual processes to understand if it can be used to train a model. Issues arise because, at the massive scale of billions of pages involved for impossibly broad generative AI projects, developers make:
No effort to verify rights or if content can be used for model training
No effort to understand if content is objectionable, illegal, unsafe, factual
No effort to check if content was generated using AI
-
Model training is the process whereby the model code (software) ingests and processes the training data to improve its ability to generate output that appears intelligent to a human user.
Model training is “self supervised” - limited human oversight or supervision
Model developers provide no transparency in model structure resulting from training
-
Deployed AI - this allows the LLM or more specialized (image/video) models to be accessed by end users.
User prompts model for response e.g. ask ChatGPT to answer a question or create an essay, ask Dall-E to generate an image based on the prompt
-
Models produce output which is probabilistic and may contain:
Hallucinations - factual inaccuracies in the model output that may be caused by the training data
Misinformation - hallucinations that replicate false claims that may be caused by the training data
Malice - objectionable, illegal and otherwise harmful content generated by a model
-
AI generated content is published across the web. From there, it’s ingested and processed by LLMs as training data. There are no LLMs that reliably determine if new training content is human or AI generated.
This process further pollutes training data, creates a circular pattern, and leads to increased potential for perpetuating misinformation and other harmful content
Dangers of no regulation
the outsized Impacts of Training Data
Training data is the foundation for which Artificial Generative Intelligence is built. As such, it has outsized impact on some of the biggest forward-looking impacts that these systems will have on public trust, safety, and consumer protection. This is why training data transparency and accountability must be addressed immediately in public policy.
Without any transparency in information sourcing, the general public will become increasingly reliant on a system that has no accountability. In addition to the risk of rampant misinformation, the mere awareness of its fallibility further erodes public trust in true information.
Threatens data privacy.
Without action, PII (personally identifiable information) via social media content or otherwise is being used without knowledge or consent. This means unregulated access to personal information and user content.
Threats to our children.
As is often the case, children are not immune to being victimized by AI. In addition to having sensitive information scraped from social media and other web inputs, there are numerous cases of pornographic images of underaged children created by AI.
Strengthens Big Tech monopolies.
Recent history, and accompanying anti-trust lawsuits, have revealed predatory, monopolistic behavior by the world’s largest Big Tech companies. Allowing unfettered access to unregulated training data, especially since they have a multi-year head start where no one was told how much training data was being ingested into their LLMs, will further compound their concentration of power because with their endless computing power and talent resources, they will be able to ingrain their AI products too deeply into society, too quickly.
It strips power & control from content creators.
Content creators and businesses are not being compensated, or recognized, for providing the input information training generative AI outputs. Although a small handful of publishers big enough to track and protect their data have licensing deals in place, most recent lawsuits filed by the New York Times and SAG members claim billions of dollars in damages from copyright infringement used solely for AGI outputs.
Spurs whack-a-mole policy.
Regulatory efforts that focus on outputs will lead to a scattershot approach that ignores the root problem: The lack of transparency over where information comes from in terms of traceability back to the input - training data. Holistic regulation including transparency and auditability of training data as an input Broader regulation is the only way to unify regulation.
timeline
AI and Policy change
← current
-
![]()
March 28, 2024
The Biden Administration releases a 34-page memo containing the rules of the road for federal agencies using AI systems.
-
![]()
March 13, 2024
The European Union approves the world’s first comprehensive AI law, the EU Artificial Intelligence Act.
-
![Information]()
December 22, 2023
The New York Times reports personal information found in model output
-
![Lawsuit]()
December 17, 2023
The New York Times sues OpenAI and Microsoft for copyright infringement
-
![legislation]()
October 30, 2023
President Biden Issues Executive Order on Safe, Secure, and Trustworthy Artificial Intelligence
-
![Research]()
October 20, 2023
A research paper is released showing that images of child abuse occur in training data for models
-
![Research]()
October 19, 2023
Stanford publishes an index ranking AI companies on their transparency
-
![Information]()
September 25, 2023
The Atlantic publishes a database of the copyrighted works (Books3) used to train AI models
-
![Product Launch]()
September 21, 2023
Microsoft launches Copilot
-
![Lawsuit]()
September 20, 2023
A group of prominent American fiction writers sues OpenAI for copyright infringement
-
![Legislation]()
September 3 - December 6, 2023
Senate Majority Leader Chuck Schumer holds AI “Insight Forums” to lay groundwork for future legislation
-
![Product Launch]()
August 22, 2023
OpenAI launches GPT3.5
-
![Lawsuit]()
July 11, 2023
A class action lawsuit is filed against Google for misusing personal information and copyrighted material to train Bard
-
![Lawsuit]()
July 9, 2023
Author Sarah Silverman sues OpenAI and Meta for copyright infringement
-
![Lawsuit]()
June 28, 2023
Authors Paul Tremblay and Mona Awad sue OpenAI for copyright infringement
-
![technical discussion]()
June 14, 2023
AI Summit London explores generative AI, scalability, and ethical AI
-
![Lawsuit]()
June 8, 2023
Mark Walters sues OpenAI for defamation after ChatGPT allegedly fabricated fake embezzlement claims
-
![Funding]()
May 23, 2023
Anthropic raises $450M in Series C funding round led by Spark Capital
-
![Funding]()
April 28, 2023
OpenAI raises $300M in Series E funding round from handful of venture capital investors
-
![Product Launch]()
March 21, 2023
Google launches Bard
-
![Product Launch]()
March 14, 2023
OpenAI launches GPT4
-
![Lawsuit]()
February 3, 2023
Getty Images sues Stability AI Inc for misusing Getty photos to train its Stable Diffusion AI image-generation system
-
![Funding]()
January 23, 2023
OpenAI raises $10B in Series E funding round from Microsoft, bringing company valuation to $27B
-
![Funding]()
April 29, 2022
Anthropic raises $580M in Series B funding round led by Sam Bankman-Fried, CEO of FTX
-
![Funding]()
May 28, 2021
Anthropic raises $124M in Series A funding round led by Jaan Tallinn, technology investor and co-founder of Skype
-
![Product Launch]()
September 22, 2020
OpenAI launches GPT3
-
![]()
January 23, 2020
Jared Kaplan paper, “Scaling Laws for Neural Language Models,” published. Sets off “arms race” for data to train LLMs.
-
![Funding]()
July 22, 2019
OpenAI raises $1B in Series D funding round from Microsoft and Matthew Brown Companies
-
![Funding]()
June 1, 2016
OpenAI raises $120K in seed funding round from Y Combinator
