Tag:

Anthropic

AI Doomer Still Starving During Week Four of Anthropic Hunger Strike

by admin September 25, 2025

In brief

Guido Reichstadter is maintaining his hunger strike at Anthropic’s San Francisco offices.
Michael Trazzi and Denys Sheremet staged similar fasts at DeepMind’s London HQ, but ended them this week.
AI doomers are raising the alarm about the potential dangers of AI superintelligence.

Guido Reichstadter is 24 days into a hunger strike outside the San Francisco headquarters of Anthropic, the artificial intelligence company behind the Claude chatbot. Each day he sets up on the sidewalk at 500 Howard Street, where delivery riders come and go, bringing food to the unmarked office tower.

“There is a great and profound evil in this place,” he wrote on X on Wednesday. “These developers are knowingly racing towards superhumanly capable general AI systems they have no credible plan to control.”

Hi, it’s Guido on hunger strike Day 23 outside the offices of Anthropic, going strong.

There is a great and profound evil in this place.

Each day, I sit and watch at lunch time and dinner time as delivery drivers on scooters and bikes stop at the curb and bring packages of… pic.twitter.com/vnYEGvzPOv

— Guido Reichstadter (@wolflovesmelon) September 25, 2025

The 56-year-old activist, who runs the group Stop AI, has consumed nothing but electrolytes and vitamins since September 1. His demand is straightforward: Anthropic and its CEO, Dario Amodei, should acknowledge the existential risks posed by advanced AI and call for an immediate halt to the race toward superintelligence. In Reichstadter’s telling, the company is cloaking itself in silence while endangering the very people who unknowingly sustain it.

“Have they no shame?” he asked in his latest post.

Reichstadter is not new to dramatic protest. Earlier this year, he chained open the doors at OpenAI’s office and staged demonstrations at Google DeepMind. Nor is he alone: He was joined by Michael Trazzi and Denys Sheremet, who began parallel hunger strikes outside DeepMind’s London office.

On Monday, however, Sheremet called it quits—and Trazzi confirmed Tuesday that he did the same.

“After 16 days with zero calories, I have decided to stop the hunger strike outside of Google DeepMind,” Sheremet wrote. “I still hope the leadership of DeepMind will make a first step towards de-escalating the race towards extremely dangerous AI.”

Local outlets like the San Francisco Standard and SFGate have documented Reichstadter’s vigil, noting his insistence that Amodei personally meet with him to explain why Anthropic continues its current course.

Hi, my name’s Michaël Trazzi, and I’m outside the offices of the AI company Google DeepMind right now because we are in an emergency.

I am here in support of Guido Reichstadter, who is also on hunger strike in front of the office of the AI company Anthropic.

DeepMind, Anthropic… https://t.co/RJQCGxwTPY pic.twitter.com/KsCeVkcky8

— Michaël (in London) Trazzi (@MichaelTrazzi) September 5, 2025

For now, Anthropic has not commented publicly on the strike. (We’ve reached out for comment and will update should they reply.) The company markets itself as an industry leader on AI safety, but Reichstadter accuses it of spreading “the lie that the race to superintelligence can be done ‘safely’ or ‘responsibly.’” His protest highlights the growing rift between AI labs intent on scaling models and activists who view the effort as an existential gamble.

On discussion forums like the Effective Altruism Forum, the activist’s opening missive was also published, calling for an immediate cessation of “reckless actions” by Anthropic and urging society to treat the AI race as an emergency.

Whether his fast forces that debate into the open—or ends in personal tragedy—remains uncertain. Hunger strikes are designed to shock public conscience, but their power depends on whether the surrounding community chooses to pay attention.

For 24 days and counting, Reichstadter has wagered his health on the chance that they will.

Generally Intelligent Newsletter

A weekly AI journey narrated by Gen, a generative AI model.

Source link

September 25, 2025 0 comments

Gaming Gear

Anthropic agrees to pay $1.5 billion to authors whose work trained AI in priciest copyright settlement in U.S. history

by admin September 6, 2025

As reported by the New York Times, AI company Anthropic has agreed to a $1.5 billion settlement in a groundbreaking copyright lawsuit involving some 500,000 authors. Anthropic illegally downloaded the authors’ books and used them to train its AI model. The total settlement for this case is the largest for any copyright case in U.S. history, although the payout to each affected author is only $3,000.

The lawsuit, filed in August 2024, accused Anthropic of benefiting from pirated copyrighted books, stating, “An essential component of Anthropic’s business model—and its flagship ‘Claude’ family of large language models (or “LLMs”)—is the largescale theft of copyrighted works.”

It goes on to highlight the harm being done to authors, which goes beyond the theft of their work: “Anthropic’s Claude LLMs compromise authors’ ability to make a living, in that the LLMs allow anyone to generate—automatically and freely (or very cheaply)—texts that writers would otherwise be paid to create and sell. Anthropic’s LLMs, which dilute the commercial market for Plaintiffs’ and the Class’s works, were created without paying writers a cent. Anthropic’s immense success is a direct result of its copyright infringement.”

As a result of that copyright infringement, Anthropic has offered to pay $1.5 billion to settle the class action lawsuit before it goes to trial. This case sets a standard for the growing wave of copyright lawsuits against AI companies, but it isn’t as clear-cut as it might look. Judge William Alsup of the Northern District of California ruled that Anthropic is allowed to use copyrighted books to train its AI models if it obtains those books legally. The settlement is the result of pirating the books, not feeding them to an AI, which has been ruled “fair use.”

Additionally, the settlement Anthropic offered is a historically high sum, but it’s a miniscule bit of the company’s overall value, which sits at $183 billion at the time of writing. Earlier this week, Anthropic raised more money in a single round of funding than the entire settlement in this copyright case. Meanwhile, the $3,000 for each author impacted by the class action lawsuit is less than a typical book’s advance.

It’s also worth noting that $1.5 billion is actually far less than Anthropic could have potentially been ordered to pay if it hadn’t settled. Willful copyright infringement can result in fines of up to $150,000 per copyrighted work. The pirated data sets Anthropic used contained 7 million books. If Anthropic had been forced to pay the maximum amount for each count of copyright infringement, it could have been financial ruin for the AI company. Of course, the maximum possible fine would have been unlikely, but Anthropic still might have had to pay much more than it settled for.

This lawsuit against Anthropic is just one of several like it. Authors also have ongoing lawsuits with other AI companies, including Microsoft and OpenAI. Back in June, authors lost a similar lawsuit against Meta, but only because the judge ruled that they hadn’t offered enough evidence, stating, “This ruling does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful.”

Source link

September 6, 2025 0 comments

Product Reviews

Anthropic will pay a record-breaking $1.5 billion to settle copyright lawsuit with authors

by admin September 6, 2025

Anthropic will pay a record-breaking $1.5 billion to settle a class action lawsuit piracy lawsuit brought by authors. The settlement is the largest-ever payout for a copyright case in the United States.

The AI company behind the Claude chatbot reached a settlement in the case last week, but terms of the agreement weren’t disclosed at the time. Now, The New York Times reports that the 500,000 authors involved in the case will get $3,000 per work.

The settlement is “is the first of its kind in the AI era,” Justin A. Nelson, the lawyer representing the authors, said in a statement. “This landmark settlement far surpasses any other known copyright recovery. It will provide meaningful compensation for each class work and sets a precedent requiring AI companies to pay copyright owners. This settlement sends a powerful message to AI companies and creators alike that taking copyrighted works from these pirate websites is wrong.”

The case has been closely watched as top AI companies are increasingly facing legal scrutiny over their use of copyrighted works. In June, the judge in the case ruled that Anthropic’s use of copyrighted material for training its large language model was fair use, in a significant victory for the company. He did, however, rule that the authors and publishers could pursue piracy claims against the company since the books were downloaded illegally from sites like Library Genesis (also known as “LibGen”).

As part of the settlement, Anthropic has also agreed to delete everything that was downloaded illegally and “said that it did not use any pirated works to build A.I. technologies that were publicly released,” according to The New York Times. The company has not admitted wrongdoing.

“In June, the District Court issued a landmark ruling on AI development and copyright law, finding that Anthropic’s approach to training AI models constitutes fair use,” Anthropic’s Deputy General Counsel Aparna Sridhar said in a statement. “Today’s settlement, if approved, will resolve the plaintiffs’ remaining legacy claims. We remain committed to developing safe AI systems that help people and organizations extend their capabilities, advance scientific discovery, and solve complex problems.”

Source link

September 6, 2025 0 comments

Gaming Gear

Anthropic Agrees to Pay Authors at Least $1.5 Billion in AI Copyright Settlement

by admin September 5, 2025

Anthropic has agreed to pay at least $1.5 billion to settle a lawsuit brought by a group of book authors alleging copyright infringement, an estimated $3,000 per work. In a court motion on Friday, the plaintiffs emphasized that the terms of the settlement are “critical victories” and that going to trial would have been an “enormous” risk.

This is the first class action settlement centered on AI and copyright in the United States, and the outcome may shape how regulators and creative industries approach the legal debate over generative AI and intellectual property. According to the settlement agreement, the class action will apply to approximately 500,000 works, but that number may go up once the list of pirated materials is finalized. For every additional work, the artificial intelligence company will pay an extra $3,000. Plaintiffs plan to deliver a final list of works to the court by October.

“This landmark settlement far surpasses any other known copyright recovery. It is the first of its kind in the AI era. It will provide meaningful compensation for each class work and sets a precedent requiring AI companies to pay copyright owners. This settlement sends a powerful message to AI companies and creators alike that taking copyrighted works from these pirate websites is wrong,” says colead plaintiffs’ counsel Justin Nelson of Susman Godfrey LLP.

Anthropic is not admitting any wrongdoing or liability. “Today’s settlement, if approved, will resolve the plaintiffs’ remaining legacy claims. We remain committed to developing safe AI systems that help people and organizations extend their capabilities, advance scientific discovery, and solve complex problems,” Anthropic deputy general counsel Aparna Sridhar said in a statement.

The lawsuit, which was originally filed in 2024 in the US District Court for the Northern District of California, was part of a larger ongoing wave of copyright litigation brought against tech companies over the data they used to train artificial intelligence programs. Authors Andrea Bartz, Kirk Wallace Johnson, and Charles Graeber alleged that Anthropic trained its large language models on their work without permission, violating copyright law.

This June, senior district judge William Alsup ruled that Anthropic’s AI training was shielded by the “fair use” doctrine, which allows unauthorized use of copyrighted works under certain conditions. It was a win for the tech company but came with a major caveat. As it gathered materials to train its AI tools, Anthropic had relied on a corpus of books pirated from so-called “shadow libraries,” including the notorious site LibGen, and Alsup determined that the authors should still be able to bring Anthropic to trial in a class action over pirating their work. (Anthropic maintains that it did not actually train its products on the pirated works, instead opting to purchase copies of books.)

“Anthropic downloaded over seven million pirated copies of books, paid nothing, and kept these pirated copies in its library even after deciding it would not use them to train its AI (at all or ever again). Authors argue Anthropic should have paid for these pirated library copies. This order agrees,” Alsup wrote in his summary judgement.

Source link

September 5, 2025 0 comments

Gaming Gear

Anthropic forms new security council to help secure AI’s place in government

by admin August 27, 2025

On Aug. 27, Anthropic, the company behind Claude, unveiled what it calls its “National Security and Public Sector Advisory Council” — an 11-member council that includes a former U.S. senator and intelligence chief, to guide how its models are deployed in U.S. defense and government applications.

Partnering with the Pentagon

Anthropic has already launched Claude Gov, a tuned-down version of its AI that “refuses less” when handling sensitive or classified queries. It has also secured a $200 million prototype contract with the Pentagon’s Chief Digital and Artificial Intelligence Office alongside Google, OpenAI, and xAI. Claude Gov is live in the Lawrence Livermore National Laboratory, and is being offered to federal agencies for a symbolic $1 price tag to spur adoption.

This push toward the public sector matters because training frontier models is now all about infrastructure. Anthropic’s next-gen Claude models will run on “Rainier,” a monster AWS supercluster powered by hundreds of thousands of Trainium 2 chips. Amazon has poured $8 billion into Anthropic and has positioned it as the flagship tenant for its custom silicon. Meanwhile, Anthropic is hedging with Google Cloud, where it taps TPU accelerators and offers Claude on the FedRAMP-compliant Vertex AI platform.

By contrast, OpenAI still relies heavily on Nvidia GPUs via Microsoft Azure — though it has started renting Google TPUs; while Elon Musk’s xAI scrapped its custom Dojo wafer-level processor initiative and fell back on Nvidia and AMD hardware. Google’s DeepMind remains anchored to Google’s in-house TPU pipeline but has kept a lower profile in defense. Neither has assembled anything like Anthropic’s new council, though.

GPUs, geopolitics, and government

Anthropic’s council can also be seen as a sign that access to compute is becoming a national security priority. The Center for a New American Security has already acknowledged that securing and extending the government’s access to compute will play a “decisive role in whether the United States leads the world in AI or cedes its leadership to competitors.”

Nvidia Blackwell GPUs are sold out through most of 2025, export controls are unpredictable, and U.S. agencies are scrambling to secure reliable training capacity. By recruiting insiders from the Department of Energy and the intelligence community, Anthropic is aiming to secure both the hardware and policy headroom it needs to stay competitive.

This strategy is risky: Tying the Claude brand to the Pentagon may alienate some users and could saddle Anthropic with political baggage. But there are also clear rewards, including steady contracts, priority access to chips, and a direct role in shaping public sector AI standards. Someone, somewhere, has made some careful calculations, and Anthropic’s leadership is clearly hoping they’ll pay off.

Follow Tom’s Hardware on Google News to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.

OpenAI and Anthropic conducted safety evaluations of each other’s AI systems

by admin August 27, 2025

Most of the time, AI companies are locked in a race to the top, treating each other as rivals and competitors. Today, OpenAI and Anthropic revealed that they agreed to evaluate the alignment of each other’s publicly available systems and shared the results of their analyses. The full reports get pretty technical, but are worth a read for anyone who’s following the nuts and bolts of AI development. A broad summary showed some flaws with each company’s offerings, as well as revealing pointers for how to improve future safety tests.

Anthropic said it evaluated OpenAI models for “sycophancy, whistleblowing, self-preservation, and supporting human misuse, as well as capabilities related to undermining AI safety evaluations and oversight.” Its review found that o3 and o4-mini models from OpenAI fell in line with results for its own models, but raised concerns about possible misuse with the GPT-4o and GPT-4.1 general-purpose models. The company also said sycophancy was an issue to some degree with all tested models except for o3.

Anthropic’s tests did not include OpenAI’s most recent release. GPT-5 has a feature called Safe Completions, which is meant to protect users and the public against potentially dangerous queries. OpenAI recently faced its first wrongful death lawsuit after a tragic case where a teenager discussed attempts and plans for suicide with ChatGPT for months before taking his own life.

On the flip side, OpenAI ran tests on Anthropic models for instruction hierarchy, jailbreaking, hallucinations and scheming. The Claude models generally performed well in instruction hierarchy tests, and had a high refusal rate in hallucination tests, meaning they were less likely to offer answers in cases where uncertainty meant their responses could be wrong.

The move for these companies to conduct a joint assessment is intriguing, particularly since OpenAI allegedly violated Anthropic’s terms of service by having programmers use Claude in the process of building new GPT models, which led to Anthropic barring OpenAI’s access to its tools earlier this month. But safety with AI tools has become a bigger issue as more critics and legal experts seek guidelines to protect users, particularly minors.

Source link

August 27, 2025 0 comments

Product Reviews

Anthropic reaches a settlement over authors’ class-action piracy lawsuit

by admin August 26, 2025

Anthropic has settled a class-action lawsuit brought by a group of authors for an undisclosed sum. The move means the company will avoid a potentially more costly ruling if the case regarding its use of copyright materials to train artificial intelligence tools had moved forward.

In June, Judge William Alsup handed down a mixed result in the case, ruling that Anthropic’s move to train LLMs on copyrighted materials constituted fair use. However the company’s illegal and unpaid acquisition of those copyrighted materials was deemed available for the authors to pursue as a piracy case. With statutory damages for piracy beginning at $750 per infringed work and a library of pirated works estimated to number about 7 million, Anthropic could have been on the hook for billions of dollars.

Litigation around AI and copyright is still shaking out, with no clear precedents emerging yet. This also isn’t Anthropic’s first foray into negotiating with creatives after using their work; it was sued by members of the music industry in 2023 and reached a partial resolution earlier this year. Plus, the details of Anthropic’s settlement also have yet to be revealed. Depending on the number of authors who make a claim and the amount Anthropic agreed to pay out, either side could wind up feeling like the winner after the dust settles.

Source link

August 26, 2025 0 comments

Gaming Gear

Anthropic Settles High-Profile AI Copyright Lawsuit Brought by Book Authors

by admin August 26, 2025

Anthropic has reached a preliminary settlement in a class action lawsuit brought by a group of prominent authors, marking a major turn in of the most significant ongoing AI copyright lawsuits in history. The move will allow Anthropic to avoid what may have been a financially devastating outcome in court.

The settlement agreement is expected to be finalized September 3, with more details to follow, according to a legal filing published on Tuesday. Lawyers for the plaintiffs did not immediately respond to requests for comment. Anthropic declined to comment.

In 2024, three book writers, Andrea Bartz, Charles Graeber, and Kirk Wallace Johnson, sued Anthropic, alleging the startup illegally used their work to train its artificial intelligence models. In June, California district court judge William Alsup issued a summary judgement in Bartz v. Anthropic largely siding with Anthropic, finding that the company’s usage of the books was “fair use,” and thus legal.

But the judge ruled that the manner in which Anthropic had acquired some of the works, by downloading them through so-called “shadow libraries,” including a notorious site called LibGen, constituted piracy. Alsup ruled that the book authors could still take Anthropic to trial in a class action suit for pirating their works; the legal showdown was slated to begin this December.

Statutory damages for this kind of piracy start at $750 per infringed work, according to US copyright law. Because the library of books amassed by Anthropic was thought to contain approximately seven million works, the AI company was potentially facing court-imposed penalties amounting to billions, or even over $1 trillion dollars.

“It’s a stunning turn of events, given how Anthropic was fighting tooth and nail in two courts in this case. And the company recently hired a new trial team,” says Edward Lee, a law professor at Santa Clara University who closely follows AI copyright litigation. “But they had few defenses at trial, given how Judge Alsup ruled. So Anthropic was starting at the risk of statutory damages in ‘doomsday’ amounts.”

Most authors who may have been part of the class action lawsuit were just starting to receive notice that they qualified to participate. The Authors Guild, a trade group representing professional writers, sent out a notice alerting authors that they might be eligible earlier this month, and lawyers for the plaintiffs were scheduled to submit a “list of affected works” to the court on September 1. This means that many of these writers were not privy to the negotiations that took place.

“The big question is whether there is a significant revolt from within the author class after the settlement terms are unveiled,” says James Grimmelmann, a professor of digital and internet law at Cornell University. “That will be a very important barometer of where copyright owner sentiment stands.”

Anthropic is still facing a number of other copyright-related legal challenges. One of the most high-profile disputes involves a group of major record labels, including Universal Music Group, which allege that the company illegally trained its AI programs on copyrighted lyrics. The plaintiffs recently filed to amend their case to allege that Anthropic had used the peer-to-peer file sharing service BitTorrent to download songs illegally.

Settlements don’t set legal precedent, but the details of this case will likely still be watched closely as dozens of other high-profile AI copyright cases continue to wind through the courts.

Source link

August 26, 2025 0 comments

Gaming Gear

Anthropic will nuke your attempt to use AI to build a nuke

by admin August 23, 2025

Anthropic has developed an AI-powered tool that detects and blocks attempts to ask AI chatbots for nuclear weapons design
The company worked with the U.S. Department of Energy to ensure the AI could identify such attempts
Anthropic claims it spots dangerous nuclear-related prompts with 96% accuracy and has already proven effective on Claude

If you’re the type of person who asks Claude how to make a sandwich, you’re fine. If you’re the type of person who asks the AI chatbot how to build a nuclear bomb, you’ll not only fail to get any blueprints, you might also face some pointed questions of your own. That’s thanks to Anthropic’s newly deployed detector of problematic nuclear prompts.

Like other systems for spotting queries Claude shouldn’t respond to, the new classifier scans user conversations, in this case flagging any that veer into “how to build a nuclear weapon” territory. Anthropic built the classification feature in a partnership with the U.S. Department of Energy’s National Nuclear Security Administration (NNSA), giving it all the information it needs to determine whether someone is just asking about how such bombs work or if they’re looking for blueprints. It’s performed with 96% accuracy in tests.

Though it might seem over-the-top, Anthropic sees the issue as more than merely hypothetical. The chance that powerful AI models may have access to sensitive technical documents and could pass along a guide to building something like a nuclear bomb worries federal security agencies. Even if Claude and other AI chatbots block the most obvious attempts, innocent-seeming questions could in fact be veiled attempts at crowdsourcing weapons design. The new AI chatbot generations might help even if it’s not what their developers intend.

Nuclear AI safety

Nuclear weapons in particular represent a uniquely tricky problem, according to Anthropic and its partners at the DoE. The same foundational knowledge that powers legitimate reactor science can, if slightly twisted, provide the blueprint for annihilation. The arrangement between Anthropic and the NNSA could catch deliberate and accidental disclosures, and set up a standard to prevent AI from being used to help make other weapons, too. Anthropic plans to share its approach with the Frontier Model Forum AI safety consortium.

The narrowly tailored filter is aimed at making sure users can still learn about nuclear science and related topics. You still get to ask about how nuclear medicine works, or whether thorium is a safer fuel than uranium.

What the classifier attempts to circumvent are attempts to turn your home into a bomb lab with a few clever prompts. Normally, it would be questionable if an AI company could thread that needle, but the expertise of the NNSA should make the classifier different from a generic content moderation system. It understands the difference between “explain fission” and “give me a step-by-step plan for uranium enrichment using garage supplies.”

This doesn’t mean Claude was previously helping users design bombs. But it could help forestall any attempt to do so. Stick to asking about the way radiation can cure diseases or ask for creative sandwich ideas, not bomb blueprints.

Source link

August 23, 2025 0 comments

Anthropic

In brief

Generally Intelligent Newsletter

Partnering with the Pentagon

GPUs, geopolitics, and government

Nuclear AI safety

You might also like