Two stories about the Claude maker Anthropic broke on Tuesday that, when combined, arguably paint a chilling picture. First, US Defense Secretary Pete Hegseth is reportedly pressuring Anthropic to yield its AI safeguards and give the military unrestrained access to its Claude AI chatbot. The company then chose the same day that the Hegseth news broke to drop its centerpiece safety pledge.

On Tuesday, Anthropic said it was modifying its Responsible Scaling Policy (RSP) to lower safety guardrails. Up until now, the company's core pledge has been to stop training new AI models unless specific safety guidelines can be guaranteed in advance. This policy, which set hard tripwires to halt development, was a big part of Anthropic's pitch to businesses and consumers.

“Two and a half years later, our honest assessment is that some parts of this theory of change have played out as we hoped, but others have not,” Anthropic wrote. Now, its updated policy approaches safety relatively, rather than with strict red lines.

Anthropic's quotes in an interview with Time sound reasonable enough in a vacuum. "We felt that it wouldn't actually help anyone for us to stop training AI models," Jared Kaplan, Anthropic's chief science officer, told Time. "We didn't really feel, with the rapid advance of AI, that it made sense for us to make unilateral commitments… if competitors are blazing ahead."

David Dee Delgado via Getty Images

But you could also read those quotes as the latest example of a hot startup’s ethics becoming grayer as its valuation rises. (Remember Google’s old “Don’t be evil” mantra that it later removed from its code of conduct?) The latest versions of Claude have drawn widespread praise, especially in coding. In February, Anthropic raised $30 billion in new investments. It now has a valuation of $380 billion. (Speaking of the competition Kaplan referred to, rival OpenAI is currently valued at over $850 billion.)

In place of Anthropic's previous tripwires, it will implement new "Risk Reports" and "Frontier Safety Roadmaps." These disclosure models are designed to provide transparency to the public in place of those hard lines in the sand.

Anthropic says the change was motivated by a "collective action problem" stemming from the competitive AI landscape and the US's anti-regulatory approach. "If one AI developer paused development to implement safety measures while others moved forward training and deploying AI systems without strong mitigations, that could result in a world that is less safe," the new RSP reads. "The developers with the weakest protections would set the pace, and responsible developers would lose their ability to do safety research and advance the public benefit."

AAron Ontiveroz via Getty Images

Neither Anthropic's announcement nor the Time exclusive mentions the elephant in the room: the Pentagon's pressure campaign. On Tuesday, Axios reported that Hegseth told Anthropic CEO Dario Amodei that the company has until Friday to give the military unfettered access to its AI model or face penalties. The company has reportedly offered to adopt its usage policies for the Pentagon. However, it wouldn't allow its model to be used for the mass surveillance of Americans or weapons that fire without human involvement.

If Anthropic doesn't relent, experts say its best bet would be legal action. But will the Pentagon's proposed penalties be enough to scare a profit-driven startup into compliance? Hegseths' threats reportedly include invoking the Defense Production Act, which gives the president authority to direct private companies prioritize certain contracts in the name of national defense. The military could also sever its contract with Anthropic and designate it as a supply chain risk. That would force other companies working with the Pentagon to certify that Claude isn't included in their workflows.

Claude is the only AI model currently used for the military's most sensitive work. "The only reason we're still talking to these people is we need them and we need them now,” a defense official told Axios. “The problem for these guys is they are that good." Claude was reportedly used in the Maduro raid in Venezuela, a topic Amodei is said to have raised with its partner Palantir.

Time's story about the new RSP included reactions from a nonprofit director focused on AI risks. Chris Painter, director of METR, described the changes as both understandable and perhaps an ill omen. "I like the emphasis on transparent risk reporting and publicly verifiable safety roadmaps," he said. However, he also raised concerns that the more flexible RSP could lead to a "frog-boiling" effect. In other words, when safety becomes a gray area, a seemingly never-ending series of rationalizations could take the company down the very dark path it once condemned.

Painter said the new RSP shows that Anthropic "believes it needs to shift into triage mode with its safety plans, because methods to assess and mitigate risk are not keeping up with the pace of capabilities. This is more evidence that society is not prepared for the potential catastrophic risks posed by AI."

This article originally appeared on Engadget at https://www.engadget.com/ai/anthropic-weakens-its-safety-pledge-in-the-wake-of-the-pentagons-pressure-campaign-183436413.html?src=rss

©

Uber is one step closer to going airborne. On Wednesday, the company previewed its air taxi booking service ahead of an expected launch in Dubai later this year. The inaugural Uber Air program will let travelers book Joby Aviation's electric air taxis through a familiar process in the Uber app.

The experience of booking an air taxi will be much like reserving a four-wheeled Uber. In the app, after entering your destination, Uber Air will appear as an option for eligible routes. The Uber app will book a flight and an Uber Black to pick you up and drop you off at a Joby "vertiport."

Uber

Joby's air taxis, built exclusively for city travel, can accommodate up to four passengers and luggage. (Uber says size and weight guidelines will be announced closer to launch.) The interior is about the size of an SUV and has "comfortable seating" with panoramic windows. They can travel up to 200 mph and have a range of up to 100 miles. Four battery packs and a triple-redundant flight computer are onboard for safety purposes.

The air taxis aren't (yet) autonomous and will each have a human pilot onboard. That would at least suggest high prices. After all, pilots aren't nearly as cheap as Uber's legion of independent-contractor drivers. But the company insists its air taxi rides will somehow be around as expensive as an Uber Black trip.

Joby

Dubai is only the beginning of the companies’ plans. The US-based Joby says it's in the final stage of FAA type certification and hopes to launch service in New York and Los Angeles. Globally, it's targeting the UK and Japan as well.

As for how realistic a US launch is anytime soon, well, that's up for debate. On one hand, President Trump signed executive orders last year that would create a pilot program to test such aircraft. But safety and cost considerations may require a grounding of expectations.

Joby

In November, Robert Ditchey, a Los Angeles-based aviation expert and test pilot, told NBC News that he didn't think air taxi service "was ever going to happen" in American cities. "They're dangerous," he warned. "We have had helicopters fail and crash on top of buildings in Los Angeles. We've had helicopters fail at takeoff and landing in airports. They're dangerous not from a fire point of view but in terms of landing on top of people and buildings." In addition, he warned that air taxis can't be developed in sufficient numbers to make them economically viable "unless they are subsidized by a government."

Uber and Joby have partnered since 2019. In 2021, Joby bought the Uber Elevate ride-hailing division, which essentially integrated the companies’ services. Last year, Joby acquired Blade Air Mobility's passenger business, which could open the door to eventually electrifying Blade's routes.

The video below shows one of Joby’s air taxis taking a test flight in Dubai.

This article originally appeared on Engadget at https://www.engadget.com/transportation/uber-previews-its-dubai-air-taxi-service-130000603.html?src=rss

©

A common theme in online age verification laws is the tension between user privacy and preventing children from accessing harmful or inappropriate content. Now the UK is sending a not-so-subtle message to Reddit on the subject, to the tune of £14.5m ($19.6 million). The nation's Information Commissioner's Office (ICO) accused the company of using children’s data and potentially exposing them to inappropriate content.

“Children under 13 had their personal information collected and used in ways they could not understand, consent to or control,” UK Information Commissioner John Edwards wrote in a statement. “That left them potentially exposed to content they should not have seen. This is unacceptable and has resulted in today’s fine.”

In July 2025, Reddit began requiring age verification to access adult content in the UK, in compliance with the Online Safety Act. However, that's only used to block under-18 users from sexually explicit, violent or other mature posts. The platform also prohibits users under 13 from accessing it altogether — and enforcement of that policy is lax. It merely requires users to declare, when signing up, that they're over 13. The ICO (accurately) described the method as "easy to bypass."

In its defense, Reddit told the BBC that it "didn't require users to share information about their identities, regardless of age, because we are deeply committed to their privacy and safety." The company said it would appeal the decision. "The ICO's insistence that we collect more private information on every UK user is counterintuitive and at odds with our strong belief in our users' online privacy and safety," the spokesperson added.

"It's concerning that a company the size of Reddit failed in its legal duty to protect the personal information of UK children," Edwards said. "Companies operating online services likely to be accessed by children have a responsibility to protect those children by ensuring they’re not exposed to risks through the way their data is used. To do this, they need to be confident they know the age of their users and have appropriate, effective age assurance measures in place.”

“Reddit failed to meet these expectations,” he added. “They must do better, and we are continuing to consider the age assurance controls now implemented by the platform.” The ICO also accused Reddit of failing to conduct a data protection impact assessment by January 2025.

The Guardian notes that the £14.5m fine is the third-largest handed down by the ICO. It trails only a £20m fine for British Airways involving a data breach disclosure and an £18.4m penalty for Marriott Hotels for exposing over 300 million customer records in a hack.

This article originally appeared on Engadget at https://www.engadget.com/social-media/reddit-fined-196-million-over-age-verification-checks-in-the-uk-173705048.html?src=rss

©