Microsoft’s Xbox blog:

Effective August 1, 2026, we will be updating prices worldwide. The price of XBOX consoles will increase by US$100 for 512 GB models and US$150 for 1 TB models. We will also be sunsetting our 2 TB model.

Last October, we increased XBOX console price by $20-$70 in the U.S. We hoped another price increase would not be necessary, and we have spent the last several months working with suppliers on options. Unfortunately, console storage and memory prices have increased by more than 2.5× and we expect another doubling by the fall of 2027. The entire consumer electronics industry is struggling with the current components crisis, but the effects are particularly hard on consoles. Unlike phones, computers, speakers, and other consumer devices, consoles are typically not sold at a profit, but instead for less than they cost to make.

I’m not offended they’re increasing prices. I’m offended only that they want people to style “Xbox” in all caps. And cry me a river regarding that “typically not sold at a profit” line they love to pull out.

What’s most telling is that Microsoft is sunsetting the high-end Xbox model with 2 TB of storage, not the low-end 512 GB one. High-end configurations typically have the highest profit margins. Not in this crisis, however. That’s similar to the way that Mac Studios with the M3 Ultra are now only available with the base RAM configuration: 96 GB. When the M3 Ultra chip debuted in March 2025, Apple offered upgrades to 256 and 512 GB of RAM for $1,600 and $4,000 respectively. Now they don’t offer those tiers of RAM at any price. The only way to buy a Mac Studio with more than 96 GB of RAM is to buy a used one — which eBay sellers are offering for $25,000 to $30,000.

Demetri Sevastopulo and Michael Acton, reporting for the Financial Times (paywalled, alas):

Apple is lobbying the Trump administration for clearance to buy memory chips from CXMT, a Chinese company that the Pentagon has put on a blacklist because of alleged connections to the People’s Liberation Army, according to six people familiar with the matter. [...]

Apple is not barred from buying chips from CXMT, or YMTC, another Chinese memory chipmaker. But the Pentagon has put both companies on its Chinese Military Company blacklist. The so-called 1260H list contains dozens of Chinese groups with alleged ties to the PLA that undermine US national security. [...]

Congress would probably object strongly if the administration blessed Apple purchases from CXMT, which is the Chinese national champion. “Apple choosing to partner with a Chinese military company would be a grave mistake,” John Moolenaar, the Republican chair of the House China committee, told the FT. “Helping the [Chinese Communist Party] succeed in its plans to dominate critical supply chains will make our country’s tech industry and economy more dependent on China at a time when we must build secure tech supply chains with our allies,” Moolenaar said. [...]

One former official warned the US risked losing another industry by letting Apple buy memory from a group that receives Chinese subsidies.

“Trump can show the courage to keep American memory alive for our security and our competitiveness or pour it down the drain so [Apple chief executive] Tim Cook can squeeze out a few more points of margin.”

In Apple’s announcement of the company’s imminent leadership transition, they said that in his new role as executive chairman, “Cook will assist with certain aspects of the company, including engaging with policymakers around the world.” It occurs to me, more and more, that Cook might be no less busy than he was as CEO.

Grace Kay and Theo Wayt, writing for the paywalled-with-no-gift-links The Information:

xAI launched an upgraded video model last week, highlighting how it’s pushing ahead with its own visual efforts even as it brings in outside help to compete with rivals in areas like coding. SpaceX also touted the popularity of its AI video tools ahead of its blockbuster IPO. What SpaceX didn’t mention, however, is that much of the consumer demand stems from Grok’s looser content rules, which have made it a major destination for generating pornography and other racy content.

Indeed, two recent xAI employees estimated that well over half of Grok’s overall traffic was driven by pornographic images and videos, adult role-play chats or other NSFW activity. On forums for Grok users, many of the most popular posts are porn. Users can generate visuals in several ways, including picking the video models through the consumer app or tapping them through other Grok products.

Maybe that’s a sustainable business model. But I don’t think it’s what SpaceX hype investors think they’re putting their money into. If they renamed the companies to SpaceXXX and xxxAI it might dampen enthusiasm for the stock, but make more clear what they’re selling.

OpenAI yesterday:

We’re beginning a limited preview of the GPT‑5.6 series: Sol, our flagship model; Terra, a balanced model for everyday work; and Luna, a fast and affordable model. Terra has competitive performance to GPT‑5.5 while being 2× cheaper and Luna brings strong capability at our lowest cost.

GPT‑5.6 Sol launches with our most robust safety stack to date. We strengthened protections for higher-risk activity, sensitive cyber requests, and repeated misuse, and spent multiple weeks finding weaknesses, pressure-testing our system, and hardening it against real-world attacks.

We believe in broad access, and we plan to make GPT‑5.6 Sol, Terra, and Luna generally available in the coming weeks. As part of our ongoing engagement with the U.S. government, we previewed our plans and the models’ capabilities ahead of today’s launch. At their request, we are starting with a limited preview for a small group of trusted partners whose participation has been shared with the government, before releasing more broadly.

Stephanie Palazzolo, reporting for The Information (and posted to X) regarding an internal Q&A hosted by CEO Sam Altman:

The reason for the staggered release, Altman explained: The federal government asked it to do so. Altman said that this was the best path for widely releasing the model as soon as possible, said one of the people. In a Thursday memo, Altman told staff that the government would be “approving access customer by customer during this preview period” for GPT 5.6. He added that he hoped there would be a more general release a “couple of weeks later” if all went well. [...]

Even so, after OpenAI had shared its plans for the limited release with top government officials earlier this week, Altman still received a call from Commerce Secretary Howard Lutnick cautioning the company against launching without receiving approvals from other agencies, according to a person familiar with the call.

It is perfectly reasonable to believe that the U.S. government should have regulatory approval over frontier AI models. It’s absurd to think this should be run by an apparatchik with zero AI expertise like Commerce Secretary Howard Lutnick.

AI regulation should be thoughtful, measured, consistent, objective, and deeply informed. It should not be impulsive, impetuous, petty, uninformed, subjective, inconsistent, and transactional. The latter, however, is what we’re getting.

Reed Albergotti and Ben Smith, reporting last night for Semaphor:

The decision, in a letter sent Friday afternoon to Anthropic, is a major de-escalation in the confrontation between the Trump Administration and one of the world’s most valuable private companies. Two weeks ago the administration imposed export controls on Mythos, leading to a shut down of the model and its cousin Fable 5 after warnings from Amazon and other companies that they could be “jailbroken” for malicious purposes.

The letter is silent on Fable 5, a weaker version of Mythos that was briefly the most powerful AI model widely available to consumers. People close to the talks said they are moving toward releasing Fable as well, though that timeline is unclear. [...]

Lutnick’s letter marks the beginnings of a new regulatory regime that gives the US government control over the release of frontier AI models.

A completely ad hoc policy of “Whatever the White House says, goes” is the makings of a terrible regulatory framework for AI. This would be true no matter who was president at the moment. But it’s particularly disastrous for this administration, which is both utterly transactional and staffed from top to bottom with anti-science know-nothings confident that their “ignorance is just as good as your knowledge”.

Howard Lutnick is making these determinations? I mean come on.