the official linkblog of the interwebs covering culture, politics, sports, and tech since 2012.
Hamilton acknowledged that Hegseth has no legal authority to bar military contractors from using Anthropic for work unrelated to the Department of Defense. When asked by Lin why Hegseth would have posted that, Hamilton said, “I don’t know.”
It’s 100% punitive. You don’t go from using the product and negotiating a new contract to expand capabilities to declaring they are a supply chain risk and can’t be used by any government contractor. It’s ridiculous.
What is the Venn diagram of people who thought Clubhouse and Sora were the next big thing, and is it a perfect circle?
I think a big "feature" of the start-up age we've been in since "apps" on our phones became a thing, is people building something because they can, not because they see a practical use for it. Then rich people give them lots of money to help shove it down peoples throats. Obviously, there has been some good to come from it, but overall, we still have this move fast and break things attitude in society where we throw something out there to see what happens without thinking about consequences or downstream effects. This, of course, is partly due to capitalism, but also because people "experience" life a lot less to know how things work in the real world.
Now I'm really starting to sound like a crotchety old man...
I just learned that friend of the Variable, Lucas, developer extraordinaire, developed the Monologue app for iOS, taking the initial concept and prototype that was vibe-coded based on the Mac app and turning it into an absolutely killer iPhone app. Still not a huge fan of the extremely AI bullish Every, but knowing that a developer like Lucas, who puts time, care, and attention into everything he does, is working on it now means this app should progress and only improve for the better. I’ve now switched to it from SuperWhisper for all my transcription needs.
Surprisingly absolutely no one with half a brain, OpenAI actually couldn’t fulfill the same deal as Anthropic while holding to the same high standards and principles.
And it’s all of these shortsighted nincompoops that are going to absolutely destroy the economy—and not just the stock market, but the everyday lives of normal folks—along with the idiotic policies, actions, and war (crimes) of the current administration.
I’m rambling and getting a bit off topic now, but we are at the point (and, to be clear, have been for sometime) in a post-capitalistic society where outputs don’t actually matter. Instead what businesses make is irrelevant as all they care about are numbers on a spreadsheet to appease Wall Street. The best ideas or products don’t prevail, just those who have inherited or “created” wealth, sitting on billions of dollars with companies that make billions in profit, but somehow can’t afford to pay full-time employees enough to keep them off of welfare.
It’s almost funny when the right wing made up this fake idea of a “welfare queen” who just wanted to sit at home and collect a check, when, in reality, they created a system where those same people have to work two jobs to survive—working harder than any CEO pencil pusher—while still sitting below the poverty line because of the actual welfare class that was created due out of Regan and trickle down economics.
“Two of our most important safety principles are prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems,” Mr. Altman said. “The DoW agrees with these principles, reflects them in law and policy, and we put them into our agreement.”
From one liar to another—so you know it has to be true. They somehow have the same principles as Anthropic, but magically came to an agreement despite those principles being the only thing killing the Anthropic agreement. More likely, OpenAI has been chomping at the bit to get at that military industrial complex complex money so there’s actual some money coming in that they can point at for their investors to oooh and ahhh at and continue to pretend they are somehow worth over $700 million.
The device had already been on the market for about three years. Until then, the U.S. Food and Drug Administration had received unconfirmed reports of seven instances in which the device malfunctioned and another report of a patient injury. Since AI was added to the device, the FDA has received unconfirmed reports of at least 100 malfunctions and adverse events.
Medical devices, including those used in surgeries and critical monitoring should never have AI because we know AI hallucinates and will always make avoidable mistakes. It may be useful for research and diagnosis (or secondary monitoring) where a human has time to review and confirm, but when someone’s life is on the line, we don’t need to add additional variables that might ignore critical information or take circuitous routes for no good reason, compounding the potential risk of human error that already exists—which AI is not reducing.
Status is the definitive nightly briefing that informs readers about what is really happening in the halls of media power.
A newsletter about the history behind today’s politics. Click to read Letters from an American, by Heather Cox Richardson, a Substack publication with millions of subscribers.
Jason Kottke’s daily links and commentary on art, science, technology, design, film, and culture since 1998.
