Anthropic putting 5% of shares toward public benefit fund that could address AI impact on workers

AI stocks bounced back Monday as investors bought into tech companies again after last week’s sharp losses, but questions remain about the cost of the AI boom.

The Nasdaq climbed more than 1%, while chip stocks snapped an eight-day losing streak, their longest in 18 years, according to FOX Business. The S&P 500 also rose, while the Dow struggled.

Behind the rebound is a growing concern about how companies are paying for AI. Much of the money going toward chips and data centers is borrowed, and rising interest rates make those loans more expensive, The Associated Press reported.

CoreWeave, which rents out computing power for AI, has become a focus of those worries because of its $50 billion debt load, according to FOX Business.

Investors will get a closer look at whether the spending is paying off when Meta and Tesla report their latest financial results Wednesday. Chip equipment maker Lam Research and data storage company Western Digital also report this week.

For now, rising oil prices and higher borrowing costs are keeping Wall Street’s comeback in check, according to FOX Business.

Anthropic did not answer a question Monday about whether the AI company would consider giving its pledged shares directly to Americans.

The company announced Sunday that it plans to commit 5% of its shares to a new Anthropic Public Benefit Fund, which could support nonprofit donations or direct distributions to specific populations.

Elon Musk urged Anthropic to take a more direct route.

“I recommend giving the shares directly to the public,” Musk wrote on X.

Fox News Digital asked Anthropic whether it would distribute any of the pledged equity directly to Americans, but the company did not immediately answer.

Musk later pledged to give one SpaceX share to every child in America next year, though he has not explained how the giveaway would work.

The State Department announced sanctions Monday against five Chinese entities linked to the alleged theft of American trade secrets, including a technology startup founded by a former Google engineer who sought to replicate the company's advanced artificial intelligence infrastructure.

The sanctions target Shanghai Zhisuan Technology Co. Ltd. and four associated investment entities under the Protecting American Intellectual Property Act, according to a State Department fact sheet.

Zhisuan was established in 2023 by Linwei Ding, who promoted the company's ability to develop technology capable of accelerating machine learning workloads and training advanced AI models using supercomputing chips.

According to the State Department, Ding applied to a Chinese startup incubator operated by venture capital firm MiraclePlus, touting his knowledge of Google's internal technology and claiming his company could replicate the tech giant's AI infrastructure for use in China.

MiraclePlus subsequently provided seed funding to Zhisuan in exchange for an ownership stake.

The State Department determined that Zhisuan knowingly engaged in or benefited from significant theft of U.S. trade secrets, while MiraclePlus provided significant support connected to the theft.

Three additional investment entities affiliated with MiraclePlus were also designated.

The sanctions were imposed under a 2023 law targeting foreign individuals and companies involved in significant theft of American intellectual property that threatens U.S. national security, foreign policy or economic interests.

President Donald Trump claimed Monday that Chinese President Xi Jinping praised America's technological progress during a recent visit, as Trump declared the United States is "substantially ahead of China" in artificial intelligence.

Speaking during a Rose Garden executive order signing focused on women's health, Trump touted America's leadership in advanced AI technology, which he prefers to call "super intelligence."

"We're way ahead of any other nation. We're way ahead of China, substantially ahead of China," Trump said.

The president then recounted what he described as praise from Xi during a visit to the United States last week.

"When President Xi was here a week ago, he said, you guys are doing very well," Trump said. "He came off the plane. He said, wow, what a job."

Trump also argued that the term "artificial intelligence" fails to accurately describe the technology.

"It's not artificial. Artificial is fake. It's super," he said, adding that the technology must be watched carefully.

The remarks came as Trump and first lady Melania Trump participated in an executive order signing establishing a new Women's Health Division at the National Institutes of Health and a White House Women's Health Task Force aimed at advancing research and treatment for conditions affecting women. 

A single hacker used artificial intelligence agents to break into websites belonging to more than 100 companies for roughly $5,000, demonstrating the surprisingly low cost of AI-powered cyberattacks, according to a cybersecurity researcher.

The Chinese-speaking hacker allegedly relied almost entirely on autonomous AI agents to probe websites for vulnerabilities and carry out intrusions, Snorre Fagerland, a cybersecurity researcher at NorForge who monitored the activity, told Reuters in an exclusive report published Monday.

Fagerland estimated the operation cost just $50 per successful breach in AI computing credits.

The agents attempted to break into more than 1,200 websites across the United States, Canada, Australia and Japan, with successful intrusions typically taking about 23 hours.

The operation reportedly targeted companies including American manufacturers Deere, SharkNinja and Cummins, and appeared to use a combination of China's DeepSeek and Anthropic's Claude Code through an open-source AI platform called Cairn.

"It's a new world now," Fagerland told Reuters.

The extent of any stolen data remains unclear. Deere said it had no indication that sensitive information was accessed, while SharkNinja said an attack on a product support website had been contained. Cummins said it was investigating a potential breach involving a third-party vendor.

The Department of Homeland Security is ordering federal civilian agencies to immediately remove Chinese artificial intelligence company DeepSeek's products and third-party services built using its technology from government IT systems, citing national security and data privacy risks.

The directive, outlined in a Federal Register notice scheduled for publication Tuesday, also prohibits federal civilian agencies from purchasing DeepSeek products and services, including those developed by affiliated entities.

DHS warned that DeepSeek's servers are located in China, where the country's legal system permits government cyber and intelligence agencies broad access to private company infrastructure.

According to the order, DeepSeek's AI systems could expose personally identifiable information and sensitive government data to Chinese authorities.

The department also warned that information collected through DeepSeek could potentially be used to develop intelligence on U.S. personnel and government programs.

The restrictions extend to third-party products and services developed using DeepSeek implementations, significantly broadening the scope of the directive beyond the company's own applications.

Federal agencies must immediately comply with the removal and procurement restrictions, although the order allows agencies to request exceptions under certain circumstances.

Anthropic's top federal affairs official is defending the company's draft artificial intelligence safety legislation following criticism over its proposed restrictions on state regulations and federal oversight of potentially dangerous AI models, emphasizing that the proposal is not final.

Anthony Cimino, Anthropic's head of federal affairs, said in a lengthy post on X that the draft circulating over the weekend was shared with a small number of congressional offices to translate the company's existing AI safety proposals into legislative language.

"We agree there are areas where the draft could be improved," Cimino wrote, addressing criticism surrounding states' regulatory authority, the government's ability to block dangerous AI models and proposed safety standards.

Cimino said Anthropic plans to revise language concerning federal preemption of state AI laws, emphasizing that states should retain authority over areas including child protection, whistleblower protections and certain AI safety requirements.

He also defended a proposal allowing the Commerce Department to seek a federal court order blocking AI models with dangerous capabilities and insufficient safeguards, while acknowledging concerns that the proposed legal standard could make intervention too difficult.

"But the bar can't be so high that the power never gets used when it's needed," Cimino wrote.

The company is also considering stronger requirements for reporting AI safety incidents and independent evaluations, including measures to prevent companies from selecting evaluators likely to give favorable assessments.

Cimino said Anthropic intends to explore additional policies in the coming months that would allow the government to adjust the pace of advanced AI development when risks warrant intervention.

"We circulated these ideas because it's important to get early feedback from policymakers," he wrote, adding that Anthropic wants its proposal to be "the strongest on the table."

Three federal agencies have issued new safety guidelines urging artificial intelligence developers to prepare for scenarios in which superintelligent AI systems act beyond human instructions, including recommendations to monitor, contain and shut down potentially dangerous activity.

The National Institute of Standards and Technology (NIST), National Security Agency (NSA) and Cybersecurity and Infrastructure Security Agency (CISA) released 10 voluntary practices Friday aimed at preventing and responding to incidents during the training and evaluation of advanced AI systems.

The guidance focuses on incidents in which autonomous AI agents take significant actions without instructions that violate security restrictions, exceed their assigned tasks or undermine human oversight.

Among the recommendations are limiting AI agents' access to sensitive systems, continuously monitoring for unexpected behavior and establishing mechanisms to rapidly block or shut down unauthorized activity.

The agencies also recommend automatically terminating AI activity when control failures are detected and potentially suspending training or testing of similar systems until safeguards can be strengthened.

The guidelines outline five stages of incident management: preparation, detection, response, recovery and reporting. They also encourage developers to notify government authorities and affected parties when significant incidents occur.

The recommendations build on existing cybersecurity practices and advance a goal outlined in the White House's 2025 Super Intelligence Action Plan to strengthen federal preparedness for risks associated with increasingly capable AI systems.

Tech entrepreneur Bill Gross is launching an artificial intelligence platform that has generated 1.2 million potential business ideas, with plans to give them away for free and help aspiring entrepreneurs turn them into companies, according to Axios.

The platform, called Idealab.ai, analyzed more than 17 million sources, including news articles, research papers, patents and company filings, to identify problems that could inspire new businesses.

Users can search the AI-generated ideas based on their interests, location, market opportunities and startup costs. Each concept includes a business brief outlining potential customers, competitors, initial steps and reasons the venture could fail.

Gross, who founded the startup incubator Idealab in the 1990s, is betting AI can help create new entrepreneurs and businesses as the technology raises concerns about job displacement.

However, Gross cautioned that AI cannot replace human judgment when making important strategic decisions.

"AI is not going to make that one for you," Gross told Axios.

The platform's first announced partner, Cloudflare, is offering $100,000 in credits to companies launched through Idealab.ai.

Gross, whose Idealab ventures have launched more than 175 companies, hopes the new platform will eventually help support 12,000 startups.

A Saudi military official paid for the wife and infant son of a former OpenAI engineer accused of secretly planting a recording device inside the artificial intelligence company's headquarters to flee the United States just 36 hours after his arrest, according to federal prosecutors.

Abdulrahman Alfozan, a Saudi national, was arrested Oct. 9, just three days before he was scheduled to fly from San Francisco to Riyadh with his family on a one-way ticket, according to a federal court memorandum filed Sunday.

Prosecutors allege that a member of the Saudi military booked and paid for a one-way, business-class flight to Doha, Qatar, for Alfozan's wife and three-month-old son on Oct. 10. The pair subsequently returned to Saudi Arabia.

The revelations came in a filing ahead of a Monday detention hearing in San Francisco, where prosecutors are seeking to keep Alfozan behind bars pending trial, arguing that he poses a serious flight risk.

Alfozan is accused of concealing a sophisticated audio recording device in an executive conference room at OpenAI, where senior company officials regularly discussed confidential technology, business strategies and government engagements.

Prosecutors allege that Alfozan used an AI chatbot to research how to build and encrypt recording devices and an OpenAI productivity tool to identify meeting rooms used by company executives.

The filing also revealed that Alfozan had been communicating with a member of the Saudi royal family who repeatedly encouraged him to join an AI initiative in Saudi Arabia.

"Saudi needs you more then [sic] the US!" the individual wrote in a May 2026 message, according to prosecutors.

Alfozan, who joined OpenAI in July 2025, was fired following his arrest. Prosecutors said he earned an annual base salary of $385,000 after a raise and received a $2.2 million equity grant when he joined the company.

Federal authorities argue that his financial resources, family ties to Saudi Arabia and apparent support from a foreign government make him a significant flight risk.

The investigation remains ongoing, and prosecutors said additional charges could follow. Alfozan currently faces a federal charge related to the unlawful interception of oral communications, which carries a maximum sentence of five years in prison if convicted.

Interpol is warning that increasingly sophisticated artificial intelligence tools are allowing scammers to convincingly impersonate friends, family members and coworkers, making it harder for people to distinguish fraudulent calls and messages from legitimate communications.

The international police organization launched a public awareness campaign Monday highlighting how criminals are using AI-generated videos, images and voices, known as "deepfakes," to trick victims into handing over money or sensitive personal information.

According to Interpol, scammers can now replicate a person's face or voice using only a small amount of content found online, allowing them to carry out convincing impersonation schemes without significant technical expertise.

"AI is erasing the tell-tale signs of a scam. Spelling mistakes, branding inconsistencies, clumsy formatting: these red flags are fast becoming a thing of the past," said Toshinobu Yasuhira, director of the Interpol Innovation Centre.

The organization warned that deepfake technology is increasingly being used to target ordinary citizens through investment fraud, romance scams, impersonation schemes and sextortion.

Examples include fake video calls from coworkers requesting bank transfers or fraudulent phone calls appearing to come from family members facing emergencies.

"Anyone can fall victim to these scams, regardless of age, region, or sector," Yasuhira warned.

As part of the campaign, Interpol released three AI-generated videos demonstrating how the technology can be exploited by criminals.

The organization urged people to independently verify suspicious requests, establish a secret "safe word" with trusted family members and avoid sending money in response to urgent or unexpected communications.

Artificial intelligence company Anthropic announced Sunday that it plans to commit 5% of its shares to a new public benefit fund that could help address AI's impact on workers as the company prepares for a potential blockbuster initial public offering.

The commitment, which is subject to stockholder approval, will go toward the newly established Anthropic Public Benefit Fund. The company said the fund could support major donations to nonprofits or direct financial distributions to certain populations.

Anthropic identified potential disruptions to the labor market as one area of focus, noting that it has already committed $200 million to studying and addressing AI's impact on workers.

"Given the pace of AI progress, bolder approaches may be needed," the company said in a blog post.

The fund's stake could also grow as Anthropic's valuation increases, with an additional 2% of company shares potentially added at the board's discretion for every 50% increase in value above a set baseline.

Anthropic said the fund will operate as a wholly owned limited liability company, with its equity prohibited from being used for lobbying or political contributions in the U.S. or abroad.

CEO Dario Amodei and Anthropic's five other cofounders have pledged to donate 80% of their equity holdings. Combined with the latest commitment and an employee equity donation program, the company said roughly 20% of its shares will be dedicated to the public good.

The announcement comes as Anthropic plans a massive IPO that could take place as soon as November.