A Washinton DC appeals court has declined to pause the US administration’s Anthropic ban, but recommended that the case be expedited.

Anthropic won its first round in court on 26 March, when a district judge granted a temporary injunction against the US administration’s decision to designate the Claude creator a ‘supply chain risk’, something normally reserved for foreign actors.

However, last night the Pentagon succeeded in a related but distinct case, as a Washington DC appeals court declined to pause the effective ‘ban’ on government use of Anthropic products. The court did, though, recognise the likely damage caused to Anthropic, and recommended the case be expedited.

The court substantially sided with the US administration in its order, saying: “In our view, the equitable balance here cuts in favour of the government. On one side is a relatively contained risk of financial harm to a single private company. On the other side is judicial management of how, and through whom, the Department of War [sic] secures vital AI technology during an active military conflict. For that reason, we deny Anthropic’s motion for a stay pending review on the merits.”

However, the court also recognised the potential harms that were being done to Anthropic and recommended the case be expedited: “Nonetheless, because Anthropic raises substantial challenges to the determination and will likely suffer some irreparable harm during the pendency of this litigation, we agree with Anthropic that substantial expedition is warranted.”

That latter request to expedite the process had been made by Anthropic’s legal team as an alternative to any stay, should that be unsuccessful, and the AI company welcomed that element of the order.

“We’re grateful the court recognised these issues need to be resolved quickly and remain confident the courts will ultimately agree that these supply chain designations were unlawful,” an Anthropic spokesperson told SiliconRepublic.com.

“While this case was necessary to protect Anthropic, our customers and our partners, our focus remains on working productively with the government to ensure all Americans benefit from safe, reliable AI.”

The judgement also found that “Anthropic’s petition raises novel and difficult questions, including what counts as a supply-chain risk under section 4713 and what qualifies as an urgent national-security interest justifying the use of truncated statutory procedures”, and that will be the fundamental question as the case proceeds.

US district judge Rita F Lin had found in the first court case, when granting a temporary injunction against the ban last month, that: “These broad measures do not appear to be directed at the government’s stated national security interests. If the concern is the integrity of the operational chain of command, the Department of War [sic] could just stop using Claude. Instead, these measures appear designed to punish Anthropic.”

It’s a view held by many. Anthropic drew the ire of the US administration after a standoff with the Pentagon, where Anthropic refused to change its safeguards related to using its AI for fully autonomous weapons, or for mass surveillance of US citizens. The relatively ethical stance in the face of huge pressure from the US administration has earned the company many defenders, and indeed a slew of new customers.

Project Glasswing

Anthropic again flexed its ethics and safety chops this week as it declined to release its powerful new Claude Mythos model to the public, as many fear the consequences of it falling into the hands of bad actors.

Instead, its Project Glasswing will bring together leading businesses, including Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JP Morgan Chase, the Linux Foundation, Microsoft, Nvidia and Palo Alto Networks, allowing them to access the Mythos preview (released on 7 April) to boost their cyber defences.

According to Anthropic, its unreleased Claude Mythos has already found thousands of high-severity vulnerabilities, including some in every major operating system and web browser.

Anthropic’s Mythos preview is significantly capable at generating exploits. In its research, the company noted that Mythos developed working exploits 181 times out of the several hundred attempts, while Opus 4.6 had a near 0pc success rate.

“AI models have reached a level of coding capability where they can surpass all but the most skilled humans at finding and exploiting software vulnerabilities,” said Anthropic, which has promised to share learnings from Project Glasswing to benefit the wider industry.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.

I will say that I think this product could be neat in a few scenarios. If you like to dye your own extensions, or you continually change the color of a streak and don’t need to buy oodles of full-size dye containers, or you’re a parent with kids that are always dip-dying their bleached ends, then it might make sense. And the resulting product does smell good. I also like the included container that lets you store leftover dye. If you’re specifically seeking washed-out, pastel tones, I still think you should just … get some hair dye and dilute it with conditioner.

ScreenshotZuvi app via Louryn Strampe

In the end, based on my assessment, the product fails to do its job. And even if this improves—even if eventually, you can get a black pigment to help darken the shades, or Zuvi moves from using primary colors to using cyan, magenta, and yellow, or the company starts making the process of bleaching your hair at home easier, or the use of generative AI stops and Zuvi starts actually standing behind its own results … at the end of the day, you’re still left with a proprietary machine that uses proprietary cartridges, all of which will become paperweights if the company folds.

I’ve combed through every single result I can find of other reviews on the Internet. I’m not a licensed cosmetologist, but I have been coloring my own hair for around 20 years, and I have yet to see a result that’s more impressive than one the user could have achieved for less money from their local beauty supply. Semipermanent fashion dyes are available in hundreds and hundreds of shades, all of which can be mixed together. It may take some trial and error to dial in your perfect pink or blue or red, but it can be done without a machine. Your local beauty supply store likely has an associate who’s been dyeing their own hair for decades and would be happy to point you in the right direction. And respectfully, if you can’t figure out color theory or add conditioner little by little to dilute your own dye, perhaps you should go to a professional and save yourself the trouble of doing it incorrectly. The Zuvi ColorBox adds more guesswork instead of taking it away.

While I do like the initial low price, color corrections from a professional are very expensive, and the refill cartridges from Zuvi (around $22 each) could easily cost the same amount as the supplies would for dyeing your hair with products from the local beauty supply, especially if your hair is long, thick, and/or curly. And I like the idea of reducing leftover dye waste, but again, you don’t need to buy a gadget for that.

“Yeah folks, it’s gonna be harder in the future to ensure OpenClaw still works with Anthropic models,” OpenClaw creator Peter Steinberger posted on X early Friday morning, along with a photo of a message from Anthropic saying his account had been suspended over “suspicious” activity.

The ban didn’t last long. A few hours later, after the post went viral, Steinberger said his account had been reinstated. Among hundreds of comments — many of them in conspiracy theory land, given that Steinberger is now employed by Anthropic rival OpenAI — was one by an Anthropic engineer. The engineer told the famed developer that Anthropic has never banned anyone for using OpenClaw and offered to help.

DJI has all but confirmed that the Osmo Pocket 4 will be unveiled later this month, and we can’t wait.

A launch event is scheduled for April 16 at 12PM GMT. The company’s teasers also suggest this will be a global release, rather than a China-only debut.

The Osmo Pocket 4 has been a long time coming. Early references to the device date back to mid-2025.

There were multiple sightings of test units surfacing in the months that followed. It’s expected to replace the popular Osmo Pocket 3. That camera helped cement DJI’s position in the compact vlogging camera space.

Muse Spark is part of a ‘ground-up overhaul’ of Meta’s AI efforts, the company said.

Nearly a year after being established, Meta Superintelligence Labs (MSL) has finally debuted its first product, a multimodal model “purpose-built” for Meta’s products.

Muse Spark is the first in the family of Muse models and represents a “ground-up overhaul” of the company’s AI efforts, Meta said in a statement. The launch comes after the company poured multiple billions into its supposed efforts towards ‘superintelligence’, a hypothetical AI system with abilities beyond human intelligence.

Muse Spark is the “first step toward a personal superintelligence”, Meta said. The model can be accessed via Meta.ai and the Meta AI app.

According to the company, Muse Spark achieves strong performance on visual STEM questions, entity recognition and localisation. It performs on par with existing models from AI rivals such as OpenAI’s GPT-5.4, Anthropic’s Opus 4.6 and Google’s Gemini 3.1 Pro.

Muse is also marketed as a way to “learn about and improve” user health, Meta added, and is expected to be rolled out to WhatsApp, Facebook, Instagram and the company’s AI glasses in the coming weeks.

The company said it collaborated with more than 1,000 physicians to curate training data that enables “factual and comprehensive” responses. For comparison, OpenAI said it worked with 260 physicians to develop its ChatGPT Health offering.

Moreover, Meta found that Muse Spark demonstrated a “strong refusal behaviour” across high-risk areas such as biological and chemical weapons. The model also does not demonstrate requisite autonomous capability or hazardous tendencies to realise threat scenarios around cybersecurity, Meta added.

Meanwhile, Anthropic’s new Claude Mythos, released in preview to select users earlier this week, was found to be significantly more capable at generating exploits than other models.

Concerned that Meta was lagging behind the likes of OpenAI and Anthropic, CEO Mark Zuckerberg set up MSL last June after acquiring Scale AI for $14.3bn and hiring its CEO Alexandr Wang to lead the team.

“This is only the start. As we expand these features, expect richer, more visual results, with Reels, photos and posts woven directly into your answers,” Meta said.

MSL has continued to make big-name hires to add to the efforts, including Moltbook founders Matt Schlicht and Ben Parr, co-founder of Safe Superintelligence Daniel Gross and Apple’s former AI lead Ruoming Pang. The company cut 600 jobs at MSL in October.

Earlier this year, Meta said that it is budgeting up to $135bn in total expenses for 2026. The growth, it said, is driven by an increased investment to support MSL as well as its core business.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.