The old sues the new; our latest "clash of titans" takes the form of one of the most significant lawsuits filed this year. Full disclosure: this post was not generated or assisted by any AI tool; it is the original work product of the author, Attorney Timothy P. Flynn.

Last week, the New York Times sued Open AI and its affiliated companies, along with Microsoft, alleging copyright infringement. The Times alleges that Open AI, in the development of its proprietary machine learning neural network, scraped millions of copyrighted Times' articles and other protected content from the Internet. 

Further, the Times alleges that despite efforts to negotiate a resolution with the defendant companies, the paper is now owed money damages for the use of its content. This claim is based on the Times' allegations that they have made a significant investment, literally over centuries, in their news platform. Here is a link to the complaint that was filed in the United States District Court in Manhattan.

Core Allegations in the NYT Complaint

The first paragraph of the Times' complaint fittingly reads like a piece of finely-tuned journalism: 

Independent journalism is vital to our democracy. It is also increasingly rare and valuable. For more than 170 years, The Times has given the world deeply reported, expert, independent journalism. Times journalists go where the story is, often at great risk and cost, to inform the public about important and pressing issues. They bear witness to conflict and disasters, provide accountability for the use of power, and illuminate truths that would otherwise go unseen. Their essential work is made possible through the efforts of a large and expensive organization that provides legal, security, and operational support, as well as editors who ensure their journalism meets the highest standards of accuracy and fairness. This work has always been important. But within a damaged information ecosystem that is awash in unreliable content, The Times’s journalism provides a service that has grown even more valuable to the public by supplying trustworthy information, news analysis, and commentary.

These are the self-proclaimed attributes the Times asserts it brings to world-class professional journalism:

• Investigative reporting; 
• Breaking news reporting; 
• Beat reporting; 
• Reviews and analysis; 
• Commentary and opinion; 
• 10.1 million digital and print subscribers worldwide; 
• 250 articles publised every day; and 
• an unparalelled archive of content

At certain points, the complaint editorializes that the cost of the world class journalism the Times brings to the news-consuming public has drastically increased due to general AI products and that flood "today's information ecosystem". The complaint notes the hundreds of newspapers that have gone out of business as a direct result of the Internet and laments the open floodgates of "misinformation".

To protect its work product, the Times alleges that it has copyrighted every edition of its newspaper for over 100 years; it has deployed a paywall; and has strict licensing agreements. 

Here are the counts of the complaint which lay out the legal theories of liability for the Defendant companies: 

Count I - Copyright Infringment

Count II - Vicarious Copyright Infringment

Counts III and IV - Contributory Copyright Infringment

Count V - Violation of the Digital Millennium Copyright Act

Count VI - Unfair Competition by Misappropriation

Count VII - Trademark Dillution

In their prayer for relief, the Times seeks statutory and compensatory damages; disgorgement; an injunction against ChatGPT; destruction of all ChatGPT models that use NYT content in violation of their copyrights, and, of course, attorney fees.

The NYT is bringing the house in this suit. All within the context of Artificial Intelligence, Artificial General Intelligence, and machine learning. 

Machine Learning Basics

The term "Artificial Intelligence" is one of the most grotesque misnomers of all time. Tech industry professionals eschew the term for the more proper: "machine learning". 

Good old fashioned AI was a complex system of math-based rules. Then, sometime around the turn of the last Century, neural computing networks -computer networks designed to function more like a human brain- began to develop along with high-capacity supercomputers, giving birth to the new era of AI or, more appropriately, machine learning. 

The idea behind machine learning is that language, through prompts, is broken down to its basic component parts -words and characters- and assigned numeric value. With massive computing capacity behind it, the machine then uses probability to determine an accurate, or humanly appropriate, output in response to a given prompt. In the coding and design process, various outputs are then ranked. Thus, through a series of prompts, the computer learns to provide a better, more responsive, higher quality output.

Another feature of modern machine learning is the large language model [LLM]. A massive amount of language data is stored in the machine from which it retrieves and fashions its natural language response to a specific set of prompts. While human users think up the prompts, its the machine that has total access to the massive LLM and vast stores of other data. Think in terms of the Library of Congress combined with every college library on the planet, and then some. 

In the NYT copyright infringment lawsuit, the newspaper alleges that when Open AI's ChatGPT program scraped all manner of language data from the Internet, it swept proprietary NYT content along with it in the process. The Times further alleges that ChatGPT favors the NYT "style" of language as it lends itself to a highly ranked quality output. Makes sense when you think about it. If ChatGPT responds to a series of prompts in the manner of a seasoned NYT journalist, the AI user is ahead of the game. 

One of the many interesting allegations contained in the NYT complaint is that ChatGPT's first two versions were constructed on open source platforms with detailed specifications made public. Not so with ChatGPT's third and fourth iterations, notes the Times. This is because, according to the newspaper, Open AI purposely concealed the data it copied from the Internet to train its latest computer models. 

As an offer of proof set forth in the complaint itself, the NYT compares ChatGPT output and the text of a NYT article; the similarity is unmistakable. Plagiarism, says the Times. In another example, the times cites to a prompt complaining that the user was "paywalled out" of a specific NYT article, and asked ChatGPT to reproduce a portion of the article. The program complied with alacrity, reproducing the copyrighted and paywalled text.  

Another very interesting offer of proof and allegation of injury is the Times assertion that ChatGPT committed what is known in AI parlance as "hallucination". Hallucination occurs when a machine, like a chatbot, generates seemingly realistic sensory experiences that do not correspond to real world input; "misinformation", says the Times. They cite to an example where the prompt seeks a reproduction of the sixth paragraph of a specific NYT article, referenced by date, title and author. The output, however, contains non-existent quotes and other text not found in the article. This has obvious implications to the Times' journalistic reputation and could lead to a claim for damages.

It will be interesting to see how Open AI and Microsoft respond to these highly specific allegations. 

What's happening over at Open AI?

You may have heard about all the drama over at Open AI when they suddenly fired their CEO, Sam Altman last fall. Open AI originally started out as a non-profit, as noted in the NYT complaint. Their stated mission back in 2015 was to develop AI for the good of humanity, not to maximize profits. The company's board of directors had a distinctly non-tech world look; mostly academics and other non-profit professionals, except for Altman, whose tech credentials are solid. 

Despite its stated mission, as the potential for this powerful computing tool came into better focus, Microsoft jumped aboard with billions of venture capital in exchange for a 49% ownership of Open AI's for profit subsidiary. Microsoft, with its myriad tech professional contacts, supported Altman's installation as CEO. 

Last November, however, a giant board of directors misunderstanding led to Altman's firing amid great backlash by Open AI's employees and by Microsoft, its benevolent investor. Some key folks at Microsoft quietly, then not so quietly, reached out to the Open AI board and reinstalled Altman. All seems to be well for the moment. Then, last week, here comes the NYT lawsuit. 

What's Next in the Lawsuit?

The Defendant companies now have the option to answer the Times' complaint or, in lieu of an answer, they can file a motion for summary judgment pursuant to the Federal Rules of Civil Procedure. 

Given the recent board of directors drama, we will stay tuned to what Open AI and Microsoft do next. They need to focus on this lawsuit because if they lose, every content generator, including this 15-year old -nearly 650 post- blog, will have their collective hands out for a portion of Open AI's profits generated from our content. 

With Manhattan as the venue, this lawsuit will feature a high tech litigation battle between some of the most sophisticated law firms in the world. Four Big Law firms representing the NYT hail from New York, Washington, DC, Seattle, and Los Angeles. 

Post #637

www.clarkstonlegal.com

Insurance litigation and government benefits law are some of the driest topics on the planet. But the two Bauserman cases decided by the Michigan Supreme Court raise the timely issue of the role of artificial intelligence in making determinations of human intent; weighing evidence; and deciding between two versions of the same events. 

Machines Determining Benefits

The problem arose when the Unemployment Insurance Agency [UIA], the government agency that processes Michigan's unemployment claims, eliminated human decisionmaking in the determination of unemployment benefits. UIA adopted a machine learning tool known as the Michigan Integrated Data Automated System (MiDAS). This program played a significant role in the Bauserman cases.

MiDAS is an automated system used by the UIA to detect fraud in unemployment claims. In 2013, the UIA implemented a new algorithm in MiDAS that automatically flagged certain claims as potentially fraudulent, triggering a review by a computer program rather than a human investigator. Their experimentation with AI was very short lived; the state stopped using MiDAS as its sole method of detecting unemployment fraud in 2015.

In just over two years, however, the damage had been done. In addition to detecting when unemployment claimants committed fraud in their applications, MiDAS also imposed quadruple penalties upon applicants it determined were intentionally fraudulent. Perhaps the straw breaking the camel's back was that MiDAS interfaced with the state and federal treasuries to claw back wrongfully paid benefits from a so-called fraudulent applicant's tax refund.

A report issued by the state's auditor general concluded that MiDAS incorrectly decided 92% of its cases. The result was that tens of thousands of applicants had their unemployment claims wrongfully denied by a machine learning program. The machine cost the state millions and led to the class action suit in Bauserman.

The Bauserman Cases

Grant Bauserman was one of the many claimants who were affected by the new MiDAS algorithm. After Bauserman filed his claim, MiDAS flagged it as potentially fraudulent based on the new algorithm; his benefits were denied.

Bauserman was fired from his job as a bartender after he got into an argument with a customer and used profanity. He applied for unemployment benefits, but his claim was denied by the state's Unemployment Insurance Agency because it determined that Bauserman was fired for misconduct.

The bartender appealed the decision, arguing that the state's definition of misconduct was too broad and could include minor infractions. The case eventually made its way to the Michigan Supreme Court, which ruled in favor of Mr. Bauserman.

In the first Bauserman case, the Michigan Supreme Court addressed when an applicant's claim must be filed in the Court of Claims. The second Bauserman case involved whether the MiDAS computerized denial of unemployment benefits impacted a constitutional right of the class members.

The court held that the state's definition of misconduct was too broad and that claimants could not be disqualified from receiving unemployment benefits unless they engaged in willful and wanton misconduct. The court also noted that the burden of proof was on the employer to show that the employee's actions rose to the level of willful and wanton misconduct. Here is a link to the Bauserman II case dealing with an state benefit applicant's constitutional rights.

Ultimately, this class action case settled last September for $20 million. Eligible claimants who had their property seized by the state can apply for compensation from the settlement fund. The settlement still needs to be approved by the trial court; the next hearing on the matter is scheduled for July.

The Nomenclature

These cases are interesting because they involve "artificial intelligence" and "machine learning". Artificial intelligence (AI) and machine learning (ML) are related but distinct concepts in the field of computer science.

AI is a broad field that encompasses many different technologies and approaches, while machine learning is a specific approach to building AI systems that relies on data-driven algorithms and statistical techniques.

AI refers to the development of computer systems that can perform tasks that would normally require human intelligence, such as recognizing speech, making decisions, and learning from experience. AI involves the development of algorithms and computational models that can reason, perceive, and understand the world in a way that is similar to human cognition. This is what MiDAS attempted to pull off.

Machine learning, on the other hand, is a subset of AI that focuses on developing algorithms that can automatically learn and improve from experience without being explicitly programmed. ML algorithms use statistical techniques to analyze data, identify patterns, and make predictions or decisions based on those patterns.

The goal of machine learning is to enable computers to learn from data and improve their performance over time, without the need for explicit instructions or intervention from a human programmer. MiDAS may have been learning from its high rate of mistakes, but financially strapped unemployed workers were being financially penalized on the machine learning curve.

The Bauserman cases are also of interest because they involve machines wrongfully affecting people's lives. The cases highlight concerns about our state government relying too-heavily on machine learning tools. The UIA's fateful decision to rely solely on a computer program to determine benefit eligibility was disasterous, harmful and expensive. 

Next Steps

This will surely not be the last attempt by governments to replace the human decision making process with computers. In fact, these cases establish the fact that governments are already identifying and implementing machine learning tools.

Until a piece of code can "get it right" at least 92% of the time, humans will continue to play a key role in the determination of state benefits. Machine interference with the process is unconstitutional says our Supreme Court.

Clarkston Legal

Post #632