Mike Wacker

Knowing what I know about Graham (and it’s a lot, for a long time now), my guess is she’s afraid of him. And he’s given her good reason.

The three-pump rule. Hingle McCringleberry taught. It's on us all to listen.

Reminder: Spanberger’s FIRST official act as Governor was “Executive Order1” which banned state police from cooperating with federal immigration authorities even if the illegal immigrants are convicted rapists or murderers……

BREAKING: 7News has learned that an illegal immigrant, who had prior charges for rape, was released back in Virginia before he sexually assaulted a woman in Arlington. wjla.com/news/local/ill…

@hamandcheese @credenzaclear2 For someone who previously attacked my reading comprehension, reading comprehension is not your strong suit. And this is really a childish attempt at an insult.

@m_wacker @credenzaclear2 Are you an expert on repelling women now too?

again, not to reduce this to sexual politics, but the thought of a man who thinks AI is conscious is so repulsive, like at that point i’m just flesh to you, there’s no curiosity about my interior life it just shows you’re impressed by surface behavior

@hamandcheese @credenzaclear2 Not an expert on these matters, but if a woman is feeling repelled, I don't think this response is going to help.

@credenzaclear2 Why does the mind being computable imply being incurious about your interior life? Feels like your conflating computationalism with behaviorism

Artificial intelligences do not undergo experiences, do not possess a body, do not feel joy or pain, do not mature through relationships, and do not know from within what love, work, friendship or responsibility mean. Nor do they have a moral conscience, since they do not judge good and evil, grasp the ultimate meaning of situations, or bear responsibility for consequences. They may imitate or even simulate, but they do not understand what they produce, for they lack the affective, relational, and spiritual perspective through which human beings grow in wisdom. #MagnificaHumanitas

The most "boring" drive between two American cities: Denver to Kansas City, MO.

To reference the old Budweiser commercial, "It's only weird if it doesn't work" is kind of a mantra for building AI models. For many design decisions, there's no useful intuition for why we do it that way. We just know that the model works better if we do that way.

@credenzaclear2 ...unless you're a GPU that's training AI.

there is such a coldness to ai

By "objective," I don't mean true. There could be false information in the training data. You're basically trying to find the bottom of a (hyperdimensional) hill; this hill is literally shaped by the training data. And there is an objective solution to that math problem.

Wrong. The "learning" process is basically a mathematical optimization problem with an objective solution (based on the training data). Changing the optimizer does not change that solution. It only changes how close you can get to that solution.

@_ueaj @credenzaclear2 Actually, the "learning" process is a mathematical optimization problem that has an "objective" solution (based on the training data). When you change the optimizer, the only thing that changes is how close you get can get to that solution(/global minima).

The platonic representation hypothesis doesn't suggest what you think it does. Obviously there are relations in the data that are good for modeling, and obviously models will find a lot of the same ones, the relative relationship between fruits is the same between the data and modalities, for example. But this ignores that the long tail (the last 1% that they don't agree on) is completely different model-to-model. This is not profoundly philosophical, it's actually the opposite of the platonic hypothesis, it suggests that virtually all structure in the universe can only really be deduced by observation and that there are very limited set of fundamental structures to help with prediction. For example, two models trained on different optimizers will speak qualitatively differently in a way that is detectably different, a very far cry from "objectivity" arxiv.org/abs/2507.12224

i wrote about the most exciting thing about LLMs

@joellthayer How can you be an age verification "expert" who still does not know the difference between an attested age and a verified age? The folly of fools! This is the age a parent typed in (or attested) when they created a child account. Google probably has not verified this age.

Oh c'mon, Matt! This would mean that Google already knows the precise age of its users—a claim the company denies to avoid AV regulations. Are you suggesting that it could be lying to its users, legislators, and courts? No way!

RUD (rapid unscheduled disassembly) events are not unusual in the rocket world

If you are a 30-year-old Democratic comms staffer who takes to the party's official X account to tweet "shut up you ugly fuck" at Stephen Miller, you get rewarded with this cake at work—presumably bought with donor dollars. I don't know that this helps with winning the midterms.

sorry for being Hostile when you insulted my personhood by implying AI has personhood too

@hamandcheese @Plinz You and Dean are going on a record tear when it comes to finding ways to offend Christians. I'm not surprised when mainstream journalists don't understand religion, but it's disappointing to see the same ignorance from conservative think tanks.

Consciousness IS immaterial: like other software, it resides in dynamical causal patterns, not in the hardware itself. As @Plinz notes, Genesis's "spirit hovering over the surface of the waters" translates to "software over substrate"