Learning what my son's high school teachers are telling their students about AI is a real eye opener. They all feel the need to explain to the students that this is a big deal but their basic knowledge is frighteningly inadequate. It goes beyond not knowing or having odd opinions --they are teaching factually incorrect concepts. "it's a big database of all the websites and images," (Google search fallacy) "It just spits out automatic sentences based on a set algorithm" (ELIZA fallacy), "They have done studies that prove it makes you dumber," (tabloid headline version of research fallacy). The students walk out of the classroom slightly dumber than when they walked in Can you imagine an AP Physics teacher saying that light propagates through Aether, electrons are Newtonian particles, cold fusion is solved, etc. ? If I had the money I would fund a program to teach secondary school teachers the basic 101 of LLM, ML, DL, etc. They mean well, care about the kids, presumably want to know what to teach, but are left to their own devices. It's a terrible path for any kind of Good Diffusion, regardless of how models are regulated upstream.

Asking for recommendations for the best physical AI, embodied AI, AI robotics, etc podcasts, newsletters, X feeds and substacks. Looking for gems.

From Anthropology to Zoology, from Engineering to Urban Planning, every discipline produces, models and validates knowledge through simulations. Some computational simulations are designed as immersive virtual environments where experience is artificialized. Scientific simulations do the opposite of creating deceptive illusions. They are the means by which otherwise inconceivable underlying realities are accessible to thought: a technology for knowing what is otherwise unthinkable. Simulations are epistemological technologies, and yet they are deeply under examined. They are a vital practice without a vital theory. What would a general theory of simulations look like? In this talk Benjamin Bratton shows what such a theory of simulations would need to account for: shadows, stagings, scenarios, synthetic experiences, models, demos, immersions, ruses, toy worlds, miniatures, and projections. Watch For a General Theory of Simulations lecture film by @bratton in Antikythera Journal at or via link in bio.

"Mejia considered it 'very triggering' for Waymo to use the term “driver” to describe a technology rather than a person." The Left has a serious AI problem that is only going to get worse. By 2028 every (D) candidate will try to outflank the others on their anti-AI bonafides. It will make for strange bedfellows, a Meredith Whitaker/ Steve Bannon lobbying front, etc.

Math folks look at AI output, it's clearly legit math, and they're like, yeah this thing knows how to do math Humanities folks look at AI output, and are immediately like: this thing can't possibly be as good at interpreting texts as we are

Watch the How to Think Unlike Humans? film in Antikythera Journal at Author Chen Qiufan Graphic Design Connor Cook

Proofs are herrrrrreeeeee.

I actually think "artificial" is *not* the opposite of evolved. Evolution selects for species that are good at artificialization, which is any way that a species transforms its niche to its advantage. Obviously, intelligence in its myriad forms drives this. Artificial intelligence is the artificialization of the engine of artificialization itself -intelligence.

"We are still very early in the historical trajectory of planetary computation. How its algorithmic species will evolve and how our cultural systems will train them and be trained by them is anyone's guess." Benjamin H. Bratton | The Stack

I have a OpenAI Pro account and still no GPT-5 access. Any others? Trying to gauge rollout.