Blog

• Comedian Sarah Silverman is one of three authors suing the company behind ChatGPT.
• They allege that the company infringed copyright by feeding the AI chatbot copies of their book.
• They are seeking statutory and other damages.

Comedian Sarah Silverman is suing the company behind ChatGPT, alleging copyright infringement of her book.

Silverman is one of three authors who are suing OpenAI, the company that created the AI chatbot, according to the court documents seen by Insider.

The three plaintiffs allege that when prompted, ChatGPT will produce a summary of their works. They claim this is copyright infringement, as they did not consent to their books being fed to ChatGPT.

ChatGPT is a generative AI model trained by ingesting vast amounts of information from websites, news articles and books, and other sources.

When prompted by users, it can produce convincingly natural responses that mimic the experience of chatting with a human.

Silverman and the other plaintiffs allege that OpenAI “benefit commercial and profit richly” from their copyrighted works and multiple other copyrighted materials.

OpenAI did not immediately respond to Insider’s request for comment, which was sent outside of normal working hours.

Court documents

The work in question is Silverman’s book “The Bedwetter,” a memoir by actress and comedian, which she owns a registered copyright.

The other plaintiffs are writer Christopher Golden, whose copyrighted books include “Ararat,” a supernatural thriller, and writer Richard Kadrey, whose copyrighted books include the dark, urban fantasy “Sandman Slim.”

They are asking for a jury trial and to be awarded statutory and other damages.

While OpenAI has never revealed what books are part of the datasets it feeds to ChatGPT, the court document alleges that many are likely to come from “shadow library” websites that illegally aggregate content that is otherwise not readily accessible.

Daniel Gervais, a law professor at Vanderbilt University, previously told Insider that he anticipates more lawsuits involving copyright law and generative AI in the future. 

The Authors Guild, a US-based advocacy group that supports the working rights of writers, published an open letter in June calling on the chief executives of Big Tech and AI companies to “obtain permission” from writers to use their copyrighted work in training generative AI programs and “compensate writers fairly.”

• The United States is providing Ukraine with cluster bombs it says have a failure rate of 2.35%.
• But past use suggests “failure rates of 10% to 30%,” according to a report prepared for Congress.
• Cluster bombs that fail to explode can pose a risk to civilians for decades to come.

The US claim that the cluster bombs it’s sending to Ukraine have a failure rate of less than 3% is unlikely to hold up under real-world conditions, with past use of the widely-banned weapon suggesting as many as 3 in 10 fail to explode.

At a press conference on Friday announcing the transfer, Colin Kahl, US undersecretary of defense for policy, said that the “dud” rate for US-provided 155-millimeter DPICMs, or dual-purpose improved conventional munitions, is just 2.35%. He compared it to the alleged failure rate for Russian cluster munitions of 30-40%.

“I’m as concerned about the humanitarian circumstance as anybody,” Kahl said. “But the worst thing for civilians in Ukraine is for Russia to win the war.”

The failure rate is important because each cluster bomb that does not explode immediately is a threat to explode later when the war is over. In Laos, for example, littered with US cluster munitions from the 1970s, thousands of people have died during peacetime from stepping on or picking up unexploded ordinance.

Most countries have banned cluster bombs, citing the long-term threat they pose to civilians. The US, Russia, and Ukraine have not.

‘Failure rates of 10% to 30%’

Russia has been criticized not just for using cluster bombs but for deploying them in densely populated civilian areas. Ukraine insists it will only use cluster bombs against military targets, with experts saying that, against entrenched Russian forces in the east of the country, they are a particularly effective substitute for the 155 mm artillery shells it’s running out of (and which the US cannot manufacture quickly enough to replace).

But as The New York Times reported, past use of the particular munitions the United States is providing suggests a far greater potential harm to civilians in the aftermath of any battle, with an Army study from 2000 indicating a failure rate of up to 14% for artillery-fired M42/46 submunitions.

The US military has said it will select from its stockpile only cluster munitions that have been assessed to have a failure rate of 2.35%. But even so, there has been a gap between past claims of the weapon’s effectiveness and actual experience.

A 2022 report from the nonpartisan Congressional Research Service noted that, despite manufacturers claiming a dud rate of just 2 to 5% for their cluster munitions, experts in real-world cleanup operations “have frequently reported failure rates of 10% to 30%.”

Defenders of the decision to send Ukraine cluster bombs argue that the weapons will ultimately save lives by pushing back a Russian occupation force accused of war crimes, including torturing and killing civilians.

In an interview with CNN on Friday, President Joe Biden said the decision to send Ukraine cluster bombs was “very difficult” but cast it as necessary given that the country is running out of alternatives, namely the artillery shells that the United States had previously provided.

“This is a war relating to munitions,” he said. “And they’re running out of that ammunition.”

Have a news tip? Email this reporter: cdavis@insider.com

• Some AI experts say we’re barreling headfirst toward the destruction of humanity.
• Some of these statements are vague and experts disagree on what exactly the main risks are.
• These are some of the potential threats from advanced AI, and how likely they are.

AI is as dangerous as nuclear war and global pandemics.

That’s according to the latest warning issued by the Center for AI Safety (CAIS). The statement is backed by major players in the AI industry, including Sam Altman, head of ChatGPT creator OpenAI. 

The warning is one of many that have been issued in recent months. Some of the tech’s early creators said we’re barreling head-first toward the destruction of humanity and others said regulation is desperately needed.

Some of these statements have left people struggling to make sense of the increasingly hyperbolic claims.

David Krueger, an AI expert and assistant professor at Cambridge University, said that while people might want concrete scenarios when it comes to the existential risk of AI, it’s still difficult to point to these with any degree of certainty.

“I’m not concerned because there is an imminent threat in the sense where I can see exactly what the threat is. But I think we don’t have a lot of time to prepare for potential upcoming threats,” he told Insider.

With that in mind, here are some of the potential issues experts are worried about.

1. An AI takeover

One of the most commonly cited risks is that AI will get out of its creator’s control.

Artificial general intelligence (AGI) refers to AI that is as smart or smarter than humans at a broad range of tasks. Current AI systems are not sentient but they are created to be humanlike. ChatGPT, for example, is built to make users feel like they are chatting with another person, said the Alan Turing Institute’s Janis Wong.

Experts are divided on how exactly to define AGI but generally agree that the potential technology presents dangers to humanity that need to be researched and regulated, Insider’s Aaron Mok reported.

Krueger said the most obvious example of these dangers is military competition between nations. “Military competition with autonomous weapons — systems that by design have the ability to affect the physical world and cause harm — it seems more clear how such systems could end up killing lots of people,” he said. 

“A total war scenario powered by AI in a future when we have advanced systems that are smarter than people, I think it’d be very likely that the systems would get out of control and might end up killing everybody as a result,” he added.

2. AI causing mass unemployment

There’s a growing consensus that AI is a threat to some jobs.

Abhishek Gupta, founder of the Montreal AI Ethics Institute, said the prospect of AI-induced job losses was the most “realistic, immediate, and perhaps pressing” existential threat.

“We need to look at the lack of purpose that people would feel at the loss of jobs en masse,” he told Insider. “The existential part of it is what are people going to do and where are they going get their purpose from?”

“That is not to say that work is everything, but it is quite a bit of our lives,” he added.

CEOs are starting to be upfront about their plans to leverage AI. IBM CEO Arvind Krishna, for example, recently announced the company would slow hiring for roles that could be replaced with AI. 

“Four or five years ago, nobody would have said anything like that statement and be taken seriously,” Gupta said of IBM.

3. AI bias

If AI systems are used to help make wider societal decisions, systematic bias can become a serious risk, experts told Insider.

There have already been several examples of bias in generative AI systems, including early versions of ChatGPT. You can read some of the shocking answers from the chatbot here. OpenAI has added more guardrails to help ChatGPT evade problematic answers from users asking the system for offensive content.

Generative AI image models can produce harmful stereotypes, according to tests run by Insider earlier this year. 

If there are instances of undetected bias in AI systems that are used to make real-world decisions, for example approving welfare benefits, that could have serious consequences, Gupta said.

The training data is often based on predominantly English language data, and funding for training other AI models with different languages is limited, according to Wong.

“So there’s a lot of people who are excluded or certain languages will be trained less well as other languages as well,” she said.

• When inflation was at its peak last summer, Michael Burry predicted it would fall within months.
• The “Big Short” investor was right, as price growth has slowed from 9.1% last June to 4% in May.
• In contrast, Burry’s warnings of a historic stock-market crash and a recession are yet to pan out.

When inflation was at its peak last year, Michael Burry predicted it would tumble within months. The bold call from the investor of “The Big Short” fame is looking mightily prescient, but a few of his other forecasts are yet to come true.

The Consumer Price Index (CPI) surged to a 40-year high of 9.1% last June. A couple of weeks before that reading was published, Burry wrote in a since-deleted tweet that he expected “disinflation in CPI later this year.”

Indeed, annualized price growth slowed to 6.5% by December. Burry returned to Twitter at the start of this year to forecast that inflation would continue falling. He’s been so right so far, as it slowed to 4% in May.

US Bureau of Labor Statistics

Burry has a penchant for making accurate, often dire predictions. For example, he called the collapse of the mid-2000s housing bubble, and his bet on a crash was chronicled in the book and movie “The Big Short.” He also flagged the risk of post-pandemic inflation as early as April 2020, when the CPI reading was below the Fed’s target rate of 2%.

However, some of the Scion Asset Management chief’s forecasts are yet to pan out. After diagnosing a historic market bubble and warning it would end with the “mother of all crashes,” he tweeted in May last year that the S&P 500 could bottom below 1,900 points, based on how the benchmark has fared during past crashes. While the stock index plunged nearly 20% in 2022, it has rallied 16% this year to around 4,400 points.

Burry also predicted last summer that American households, under pressure from rising prices and soaring borrowing costs, would exhaust their savings by the end of the year. He cautioned that could spark a decline in consumer spending and erode corporate profits.

Moreover, he tweeted in January that the US would likely fall into recession, spurring the Fed to cut interest rates and the government to stimulate the economy, causing inflation to spike once again.

For now, consumer spending and employment are proving resilient, fueling hopes that the US economy might keep growing and escape a downturn. The Fed, which has hiked rates from nearly zero to north of 5% since last spring, has also signaled it plans to raise them further.

Regardless, it’s worth paying attention to Burry, given his track record of predicting market events. The investor highlighted some of his greatest hits in a now-deleted tweet last summer:

“Just getting one thing right is hard. 1999 tech bubble, 01-05 value revival, 2005 housing bubble, 2009 almond farms, 2020 COVID bottom, 2020 lockdown horrors, 2021 meme stocks, 2021 crypto leverage, 2021 inflation, 2022 not done yet, late 2022 ????”