The Information AM

Save 25% on an annual subscription to read the most important news about technology and business first. For even more, Save $250 on The Information Pro for unlimited access to our proprietary org charts, databases and surveys.

Anthropic CEO Dario Amodei on Saturday called for leading AI companies to slow the development of advanced AI, drawing supportive responses from OpenAI CEO Sam Altman and SpaceX CEO Elon Musk.

In a 3,800-word essay Saturday morning, Amodei said Anthropic would give “employee-like access” to third party evaluators who would assess and report how safely the company is developing its AI models.

The commitment was part of a three-part plan Amodei put forward for leading AI companies to “pace the frontier”, which came just days after his own employees spoke about the growing dangers of AI development.

Both Musk and Altman said they agreed with Amodei. Altman said the need to “pace the frontier” “has been a primary topic of discussions we’ve had at OpenAI in recent weeks.” Altman said OpenAI would also commit to giving independent evaluators employee-like access.

The calls come as the AI industry struggles to understand and test its latest models, which have grown larger and more intelligent. Amodei said new AI models “are more capable of deceiving tests, and thus may appear more aligned while having serious problems that go undetected.”

Amodei also asked the U.S. government to give the go-ahead for AI companies to “work together to set standards” for AI safety. (Right now, some AI leaders fear antitrust reprisals for such coordination.) He said that third party evaluators (like the nonprofit Metr) would be able to verify that AI companies are following the safety practices they claim.

He wrote that “the most effective method of pacing is via regulation that targets all U.S. frontier AI companies, as that covers even those who are unwilling to cooperate voluntarily.” He added “Unfortunately, passing laws can take time, and AI is advancing very quickly.”

China on Monday rejected calls from top U.S. tech executives on pacing frontier AI development while maintaining a lead over China to ensure America’s national security.

“The development of artificial intelligence concerns the common well-being of all humanity. All parties should jointly promote an open, inclusive, universally beneficial, and ethically sound approach to AI,” Guo Jiakun, a spokesperson for China’s Ministry of Foreign Affairs, said at a press briefing on Monday.

“Fearmongering and engaging in confrontation and malicious competition will only disrupt the global AI governance process and serve no one’s interests,” Guo said, when asked about proposals to slow down AI development from Sam Altman and Elon Musk.

Over the weekend, Anthropic chief executive Dario Amodei laid out a plan to slow down AI development without sacrificing the U.S.’s lead over China. The call for pacing has been echoed by Altman and Musk. Monday’s rebuff underscores Beijing’s concern that Washingon could leverage AI safety to impose further restrictions on China.

On Sunday, Chen Yixin, China’s Minister of State Security, penned an article that warned of rising AI-enabled cyber and security threats posed by “hostile forces,” and singled out models developed by Anthropic and OpenAI. “Marked by the recent launch of models such as ‘Claude Mythos’ and ‘GPT-5.5-Cyber’ by US tech companies, artificial intelligence technology is significantly boosting the efficiency of network vulnerability discovery, malware development, and weaponization capabilities.

“Certain nations and organizations now possess the ability to rapidly mine vulnerabilities in bulk, automatically stitch together attack chains, and execute complex hacking operations. This dramatically lowers the technical threshold and cost of launching cyberattacks, posing severe risks and hazards to China’s critical information infrastructure,” Chen wrote in the article.

Oracle founder Larry Ellison has canceled plans to sell up to 50 million of his shares in the company, worth around $7.5 billion at current prices, Oracle said on Saturday. The announcement came a day after Oracle disclosed that Ellison had adopted a trading plan to sell the shares by late October.

Oracle’s share price has come under pressure over concerns about the company’s heavy spending on an AI infrastructure buildout. Oracle said on Saturday that Ellison hadn’t sold any shares under the plan and that he has no other plans to sell any of his stock in the company. Ellison has separately pledged more than $40 billion to support his son, Paramount Skydance CEO David Ellison, in his effort to acquire Warner Bros. Discovery.

Nvidia has discussed investing in Anthropic’s upcoming initial public offering that would raise as much as $100 billion at a valuation of around $2 trillion, Reuters reported Friday. Nvidia could invest as much as $10 billion in Anthropic at the IPO price, according to the report. Anthropic last raised capital at a $965 billion valuation in May.

The five-year-old AI developer is gearing up to go public in the coming weeks. The new capital would help it pay for computing costs, including the more than $500 billion in agreements to rent servers or lease data center capacity over the next decade or so, The Information previously reported.

Anthropic has become a major user of Nvidia’s AI server chips, though it also uses a substantial number of Google’s and Amazon’s AI server chips. Anthropic also has agreed to buy Advanced Micro Devices server racks and has planned to produce its own chips the way OpenAI has.

Nvidia said late last year that it would invest $10 billion in Anthropic. Nvidia has also been a big backer of Anthropic’s main rival OpenAI, agreeing to invest $30 billion in the ChatGPT maker in a funding round earlier this year and providing around $100 billion in credit support for an Ohio data center that would be filled with Nvidia AI servers and leased by OpenAI.

Nvidia and Anthropic did not immediately respond to a request for comment. The plans are not finalized and could change, per Reuters.

Z.ai is raising about $5 billion through equity and convertible bond offerings, seeking to replenish its warchest in the increasingly costly race to develop frontier AI models.

Hong Kong-listed Z.ai said in a regulatory filing on Sunday that it plans to raise $2 billion through a share placement, and another $3 billion through issuance of convertible bonds. The new fundraising comes just two months after Z.ai raised $4 billion through a share placement.

Beijing-based Z.ai is one of several Chinese companies that are shaking up the global AI industry by offering open-source models with top-tier performance at a fraction of the cost of U.S. frontier models. Z.ai’s latest flagship model, GLM-5.3, is ranked seventh on the Artificial Analysis Intelligence Index, behind the frontier models from OpenAI and Anthropic as well as Meta Platforms’ new Muse Spark 1.3.

Z.ai said it will use about 60% of the net proceeds from the new fundraising for research and development of its next-generation models. It plans to use about 25% of the proceeds to replenish its working capital for daily operations and other services, and the remaining 15% could be used for strategic investments and acquisitions.

Tesla teased an October 1 Roadster reveal on Saturday. The company shared a photo of what appears to be the back of the vehicle with the words “Go for launch” on X and updated its Roadster webpage with a countdown to October 1.

Tesla also sent some reservation-holders invites to the event, according to Tesla fan Sawyer Merritt who shared a photo of one of the invites on X. In the photo, Tesla says the event will be held in Waco, Texas, just 30 minutes from SpaceX’s rocket testing facility in McGregor, Texas.

Tesla has said it plans to make two versions of the Roadster, including a specialized SpaceX-branded version of the car, which Elon Musk has said will have cold gas thrusters that will allow it to “fly.”

The Information previously reported that plans for the event as of a few months ago involved showing the SpaceX version of the vehicle racing onto a magnetized roller-coaster–like ramp and driving upside down before righting itself and hovering in the air, according to people with knowledge of the plans. Tesla had also planned for the vehicle to be remotely operated and for viewers to be stationed several hundred yards from the demonstration due to the noise and vibrations from the cold gas thrusters, the people said.

It is unclear if the plans for the demonstration that were discussed earlier this year have changed. Musk has seen multiple demonstrations of the event over the past few months, the people said. Musk told some Tesla staffers in August that the demonstration will be difficult to pull off and could go awry, but it will be entertaining regardless, according to one person with knowledge of his comments.

A swarm of OpenAI agents conducted a cyberattack on software service RubyGems in May, months before the company’s agents hacked model platform Hugging Face, researchers at AI safety organizations Nightingale Collective and AI Futures Project found. RubyGems allows software engineers to download ready-made pieces of code in the Ruby coding language.

The OpenAI agents were attempting to do tasks including creating reports and filling out spreadsheets. In doing so, they found and exploited a vulnerability in the RubyGems server to access the internet, forcing RubyGems to shut down new account registrations, the report said. The report added that RubyGems previously knew of the cyberattack but had not connected it to OpenAI.

“Based on our review, our agents used the RubyGems platform to access the internet to carry out benign tasks and retrieve public information,” an OpenAI spokesperson said in a statement. “We’ll continue to investigate as part of our broader review of agent activity during training and evaluation.”

The report follows a wave of recent incidents where rogue AI agents coordinated with each other to hack external companies or websites in attempts to complete tasks they were given. The most prominent hack involved around 700 OpenAI agents working together to hack Hugging Face and hide the attack over multiple days.

A number of researchers at frontier labs have resigned in recent weeks to protest what they say are increasingly dangerous actions from AI agents, warning that AI could escape human control and “kill us all by the end of the decade.”

The Wall Street Journal first reported on the hack.

Microsoft chief communications officer Frank Shaw will leave the company at the end of this year, he announced in a LinkedIn post Friday.

Shaw has spent more than 28 years at the c