NFL teams doing schedule release videos has become a little tradition — and while the Cardinals are getting ruthlessly dunked on for doing AI slop, the Packers are making it clear theirs was all hand-made. (An increasing trend in advertising overall.) Anyway, disclosure: I am from Wisconsin, and the Bears still suck.
AI
Artificial intelligence is more a part of our lives than ever before. While some might call it hype and compare it to NFTs or 3D TVs, generative AI is causing a sea change in nearly every part of the technology industry. OpenAI’s ChatGPT is still the best-known AI chatbot around, but with Google pushing Gemini, Microsoft building Copilot, and Apple adding its Intelligence to Siri, AI is probably going to be in the spotlight for a very long time. At The Verge, we’re exploring what might be possible with AI — and a lot of the bad stuff AI does, too.
- RELATED /





Personalized health is the holy grail, but there’s a long way to go before algorithms can factor in chronic conditions.
Researchers at the security firm Calif say they used Anthropic’s cybersecurity AI to create a privilege escalation exploit, the Wall Street Journal reports:
Last September, Apple said it leveraged its hardware and operating system expertise into a technology called Memory Integrity Enforcement (MIE), which it described as “the culmination of an unprecedented design and engineering effort, spanning half a decade.” With Claude, building the code that exploited the two MacOS bugs took five days, Calif says.

Journal editors and peer reviewers are being flooded with AI-generated papers that are almost impossible to detect.
That’s according to Amazon’s President and CEO Andy Jassy — a cold, yet unsuprising take from the man who’s planning to replace 600,000 employees with robots by 2033. In this Bloomberg article, you can read more about how he’s overhauling Amazon to go all-in on AI.
The musician, producer, and long-time Taylor Swift collaborator described people who use generative AI tools as “Godless whores” in a rant on Instagram:
“To everyone who is gassed up about the new ways you can fake making art, by all means drive right off that cliff. We’re genuinely happy to see you go.”
As part of a new release, “your ChatGPT subscription can now power an OpenClaw agent that feels much closer to the model it is built on,” OpenAI’s Nik Pash says in a blog post.
The OpenClaw team has also been working “really hard on performance, reliability, security, and stability,” according to OpenClaw founder (and OpenAI employee) Peter Steinberger.
This investigation by the Florida Trib and the podcast Question Everything is a sobering read about the state of local news right now and how AI is being used to fill the void with sketchy junk.
[The Florida Trib]

The trial felt less like the fate of OpenAI and more like a window into petty grievances.
God bless. We are in the Microsoft closing statements. “Microsoft never found a single page of a single document” that referenced Musk’s alleged restrictions on his donations during the due diligence process.
Microsoft doesn’t want any of this
He may be laying it on thick, but he did get a big laugh.
He reminded the jury that Musk isn’t in the courtroom while Altman and Brockman are. (Musk posted yesterday that he was en route to Beijing on Air Force One.) “They are here because they care a lot about this,” Savitt said. “Mr. Musk isn’t here. Mr. Musk came to this court for exactly one witness — Elon Musk — and he hasn’t been seen since. Now he’s in parts unknown.”
“He claims to have heard things high atop a windy hill where no one else can hear,” Savitt told the jury. (Strange phrasing, but after the bridge metaphor from Molo, I wouldn’t expect anything less.) He also says Musk has “unclean hands” due to his “unconscionable conduct” related to the claims he’s bringing. “Only after OpenAI succeeded, against Musk’s prediction, only then did he start threatening litigation,” Savitt said.


He said that by his calculations, people said things like “I don’t remember” and “I don’t recall” between 150 and 200 times during this trial so far — using this to bolster his argument that Musk had waited too long to bring his claims.
You may remember that yesterday I was completely tickled by the possibility that the jury might get to see this. Even YGR seemed tickled by it. Unfortunately, she ruled that discussing it was fine but unless the Musk team gave them reason to introduce it, the jurors wouldn’t see so much as a photo. But this is the trophy Josh Achiam got for getting yelled at by Elon Musk.
There’s one more thing that Savitt is harping on. “Has the OpenAI nonprofit respected its general founding principles?” The question doesn’t matter, legally, since Musk didn’t create a charitable trust, but Savitt is going to spend some time on this because Molo emphasized it.
As xAI aims to compete with rivals like Anthropic and OpenAI, which have focused heavily on AI-assisted coding, the company is launching a new agentic coding CLI tool for Grok. It’s available initially for subscribers to xAI’s SuperGrok Heavy plan.
[xAI]
Musk doesn’t want to admit that trying to build an AGI lab in Tesla was a failure — whether that was by acquiring OpenAI or trying to poach all its talent, maybe even putting Altman on the board. Eddy suggests this case is revenge on OpenAI for succeeding.
That’s kind of where I’ve landed! The idea of the “adjunct” for-profit (Eddy says this is a moving target, and though Musk used it twice in testimony, when Savitt used it, Molo objected and accused him of making up a term) doesn’t show up in any of the brainstorming structure documents. We do see parallel for-profits, and the idea of a conversion to a for-profit and shutting down the nonprofit. Jared Birchall also testified that he filed to register a company for this.
They went to Ilya Sutskever and Greg Brockman, among others, right before he proposed that he get 62.5 percent of a for-profit company. We are now looking at tax forms and letters — neither of which show any specific purpose. Jared Birchall also testified that there was no specific purpose for the donations. Shivon Zilis doesn’t remember it. Sam Teller doesn’t remember it. This is like watching the Warriors play a team of 6-year-olds.
Chronology and documents. Musk’s performance on the stand does give credence to the suggestion from Eddy that Musk “took his marbles and went home” when he couldn’t get his way.
She opens with a banger. Musk has said he made donations with strings attached. “Even the mother of his children can’t back his story.”
I do wonder how this would have played in the hands of a better lawyer. Molo’s book report did not overwhelm me with confidence in his case, particularly because a lot of his point-blank assertions were profoundly arguable.
In all that chaos, Microsoft did suggest board members. But OpenAI didn’t take those suggestions — except one, well after the crisis. I don’t know man, I don’t really understand how this goes to the Microsoft case. It might be a suggestion the nonprofit board doesn’t really control the for-profit — but Helen Toner’s and Tasha McCauley kind of came off as amateurs in their approach, not least because there was no investigation before the firing.
So far he’s got “they’re a for-profit corporation,” “they know Musk was a co-founder,” and “they read the announcement OpenAI existed.” This is easily the thinnest part of a very thin case — on the OpenAI part, there’s at least Brockman’s diaries.
During his speech, Molo told the jury he wasn’t asking for money. That is in fact not true — otherwise I wouldn’t be sitting through phase 2 of the trial next week. “You slipped it in nicely,” YGR says. But Molo needs to retract that statement.


I am just going to break from telling you what Molo is saying to say what my personal impression was from sitting here all these weeks: Everyone was improvising. There was no plan. This is especially true of “the blip.” I do wonder if there’s a way to incorporate that into Musk’s case. Anyway, Molo just referenced an exhibit he didn’t have handy, asked for an exhibit number, and then said he’d get it for the jury late.r I have to say, I know Musk’s team is smaller than OpenAI’s, but this might have been a moment to call in another lawyer to handle the close. Someone who could have prepped better, perhaps. Marc Toberoff, who’s theoretically a key figure on this team, hasn’t stood up to do a single thing. Maybe this could have been his moment, I don’t know!
OpenAI thought it would get more subscribers from how ChatGPT was baked into Apple’s operating systems and that there would be deeper integration, according to Bloomberg’s Mark Gurman.
The company is considering taking legal action against Apple, which could include “sending the iPhone maker a notice alleging breach of contract without necessarily filing a full lawsuit at the outset,” Gurman reports.
Molo keeps interrupting himself to restate things or say things like “remember the residuals?” It’s really important in closing to tell a straightforward, easy-to-follow story here, because this is where you put together all the testimony into your case. I know Molo’s got an uphill battle on the facts here, but this could be smoother. He did also just call Greg Brockman “Greg Altman.”
And that the “important constraints” of the capped-profit structure in the first two Microsoft investments did not breach the charitable trust. However, the 2023 investment”changed the world. Altman, Brockman, and OpenAI breached the charitable trust created by Elon” by enriching investors and insiders at the expense of the nonprofit and not open-sourcing the technology.













