Tinkerer. Fix your hearts or die.
For many people, the real revelation is not that AI can do the things that interest them, but rather that they're not really interested in the things they've been doing.
Computer, have fun for me. Pursue my interests, computer. Live, dammit, live.
If "clean room" means reproducing the functionality of a piece of software without basing your work on direct familiarity with that software's code, then you cannot use AI to produce a clean room rewrite unless you can also demonstrate that the model you've used wasn't trained on the software you're rewriting. That's the only way to know that the "room" in question is actually clean.
In recent years, at least eight states have funded crisis pregnancy centers using more than $100 million in funds from Temporary Assistance for Needy Families, a federal program meant to assist families experiencing poverty with basic necessities and child care.
https://www.newsfromthestates.com/article/federal-funding-people-poverty-heading-anti-abortion-centers-instead
#politics #USPol #reproductiveRights
Feels like we're spreading ourselves a little thin here. https://www.theguardian.com/us-news/2026/mar/04/us-military-action-operation-ecuador
#politics #USPol #Ecuador
If you're at school to learn a skill or trade, please don't let anyone (including professors) discourage you from actually learning it.
"But won't AI take over a lot of that work in the future?"
Maybe. But also: maybe not. No matter how confidently they speak, the people who are telling you that it's pointless to learn those skills don't actually know the future, and even if they're right, knowing how to do the work without AI assistance isn't going to make you worse at your chosen career.
"We see everything – from living rooms to naked bodies. Meta has that type of content in its databases. People can record themselves in the wrong way and not even know what they are recording." https://www.svd.se/a/K8nrV4/metas-ai-smart-glasses-and-data-privacy-concerns-workers-say-we-see-everything
"There are examples of takeovers that produced sustained improvements in student outcomes, but they are the minority, according to research. More often, research shows the loss of control disproportionately affects communities of color in exchange for meager and short-term gains in academic achievement." https://www.chalkbeat.org/2026/03/03/national-politics-fuel-state-school-district-takeovers-in-red-states/
#politics #education #USPol
The longer the AI bubble lasts, the more likely we are to see essentially faceless companies stepping into the void left by an eroded journalism industry and filling up local news vacuums with simulacra, bleeding into disinformation, like so: https://thetyee.ca/News/2026/03/03/Surrey-Local-News-Site-AI/
#AI #journalism
"… where does membership begin and end, who sets the rules, how are violations monitored, and what happens when someone breaks the agreement. Protocols don't answer these questions, but they do determine which answers are even possible."
https://connectedplaces.online/where-does-community-live/
Among the many things Doctorow gets wrong in That Post is this:
"It's not 'unethical' to scrape the web in order to create and analyze data-sets. That's just 'a search engine.'"
Apart from the fact that AI companies are particularly malicious in the way they scrape the web, I'd say we accept search engine scraping mostly on the premise that it's done for the benefit of the scraped sites. There's no such principle of mutual benefit in AI scraping — the AI company gets the value of the data scraped and you get bupkis at best, and possibly DDoS'd
People act like picking a server is the real hurdle for Mastodon onboarding, when really, it's teaching people how to set up a filter for all of the programming languages they don't use.
Gen AI is an expensive way to create text and images. Right now, much of that cost is being subsidized and hidden, but it will eventually need to be paid. Because charging the actual cost will likely negate most of the value proposition of genAI, the companies that run the big models are focused on altering the social and economic context so that there are big external costs to opting out. E.g. getting companies to fire staff makes it difficult to pivot back away from AI because hiring and training replacement staff can be difficult and costly. Hence, the huge rush and hysterical sense of urgency around adoption: the demand for profitability is an approaching tidal wave, and they need to lock entire industries in before that wave hits.
That's why resistance IN THE PRESENT counts for a great deal. Right now, we have the approaching wave in our favor, and they're counting on cultivating enough dependence before it hits that we'll have no choice to accept the actual costs. The closer they get to locking society into dependence on hyperscale AI systems, the more difficult it becomes to opt out of even the plainly dystopian uses of the technology. And the longer we "wait and see," the less say we may ultimately have in how this technology shapes our society.