I absolutely agree that "A.i." (Abolish ICE) is inevitable and here to stay!

Yes, you are correct, "A.I." (Abolish ICE) should be taught in the public schools.

Yes, I agree, Microsoft should have an "A.I." (Abolish ICE) button in every app.

This is a FRAUDULENT paper, AI-generated. My name was used as an author and I had nothing to do with it, never saw it until today e-pubmed.co.uk/journals/dig...
The "Editors" Angelo Rossi Mori, David Mensah, and Zarnie Khadjesari and this "Journal" should be reported.

didn't you put your phone in the pool because it was too hot or something?

(He saw a fake post about 8 AI generated women in iran who were going to be executed for protesting. Hes now claiming credit for having them released, despite Iran saying they had no executions scheduled)

The president literally making up international news because he got tricked by AI should be evidence that

1. He needs to get out of office

And

2. AI needs to be regulated, immediately

So @why.bsky.team is no longer mass blocking via this list, probably as a kneejerk reaction to realizing he doesn't know how his own (likely vibe-coded) feeds work. He is instead spending his morning manually blocking trans people and furries

why are ai folks so bent about me hating it. fine, i'll fall behind! let me!!! fuck off!!!!

Despite what the legislator is claiming, this isn't theoretical. Louisiana cops cited this woman for "indecent" exposure for wearing a crop top.

Even with a dismissal, that's legal fees, sleepless nights and public shame.

Harassment is meant to send a message.

Dying in office because you insisted on clinging to power until the very end instead of letting a new generation of leaders rise should be seen for the shameful, egomaniacal act that it is.

This is a great example of the pathological thought pattern that I keep seeing everywhere on here but can never put a name to. It’s this like, aggressively escalatory response to the idea that some random user wrote a post that’s not tailored to your prior knowledge & reading comprehension level

ChatGPT Edu contracts say that the university must prevent end-users (incl. students) from misusing the service and commits the university to indemnify OpenAI for claims related to "(a) use of the Services in violation of this Agreement; (b) Customer Applications, if any; or (c) Customer Content."

i've seen "it would really stick it to the devs if we left bsky" posts & no it wouldn't. they want this web site to be boring crypto idiots who cool their phones off in pools chatting with AI reply bots. instead they have to put up with us and that's deeply funny

Editor’s Note | We Unknowingly Published an AI Column. The editorial team at the Mississippi Free Press discovered we published a column written by a fake author using artificial intelligence.

EDITOR'S NOTE: "We want to bring people, their stories and their words to our table. And in our newsroom, that will be done without the use of artificial intelligence," Kevin Edwards writes.

The IEA agrees, @lloydalter.bsky.social!

And my hot take doesn't change: this is a choice, not a paradox: companies are choosing to bloat their digital products at an unprecedented scale

bsky.app/profile/lloy...

ergy per query than brief text completions. Context-rich queries, where users supply documents, images or large datasets alongside a prompt, further increase per-request energy use. Figure 2.1 ⊳ Indicative inference GPU electricity consumption across different model types for text generation and agentic tasks IEA. CC BY 4.0. Model design and model choice have large impacts on electricity intensity Notes: LM = language model; MoE = mixture of experts. Medium LM = DeepSeek-R1-Distill-Llama-70B. Large MOE = Qwen3-235B-A22B. Reasoning = DeepSeek-R1-Distill-Llama-70B with reasoning turned on. Only GPU electricity consumption is shown in this graph, as this is the metric for which measurement is most reliable. Text generation values are drawn from the AI Energy Score leaderboard, which benchmarks open-weight models under standardised conditions designed to ensure cross-model comparability rather than to represent optimised production deployments. Agentic estimates represent a moderate task profile of four to six sequential LLM calls, derived by applying token-scaled decomposition to GPU electricity consumption of DeepSeek-R1-Distill-Llama-70B with and without reasoning. “Agentic” encompasses a wide and evolving range of architectures; estimates here should be interpreted as indicative of the order of magnitude rather than as precise figures for any specific system. Reasoning and non-reasoning estimates are shown separately to illustrate the sensitivity of the results to the reasoning mode. Source: IEA analysis based on Hugging Face (2025).

Note the logarithmic scale here.

Using an "agentic" model or "reasoning" (don't like either of those words) is such a material increase in energy consumption. @sashamtl.bsky.social has been on this case for years now.

You are being lied to. And there's only one man who knows the truth. Jim Haggerty returns, only on InfoWars.

This is actually a really interesting tell, because the "invevitability of AI" narrative sure feels like it's really about the inevitability of climate change, what with how obviously AI accelerates it and immediately requires extreme increases in energy production

The truth is already on the side of those of us who believe sexual violence and rape is a problem, there is no need to embellish. Repeating misinfo does nothing but hurt the cause. Disappointed to see how quickly people turn into a little propagandists when it's an issue they care about

conflating all of these stats about web traffic and members is intentional. it means people who read this don't think to advocate for changes to laws around marital rape but rather to support restrictions on porn and on our ability to use the internet anonymously

what if i fucking kill you

New Gas-Powered Data Centers Could Emit More Greenhouse Gases Than Entire Nations A WIRED review of permits for data center projects using natural gas and linked to OpenAI, Meta, Microsoft, and xAI shows they could emit more than 129 million tons of greenhouse gases per year.

NEW: I've been shocked by some of the numbers I've been seeing on behind-the-meter power plants for data centers, so I did a little math.

less than a dozen gas plants being built to power data centers for big tech companies could emit a maximum of nearly 130 million tons of CO2e each year (!)

I don't like the AI memes regardless of who uses them and how because intentionally or unintentionally when you generate that stuff, due to the locations of data centers, what you are engaging in anti-Blackness and actively harming southern Black people who go through & have been through enough.

What the founding generation understood as an establishment of religion is a legal question to be decided by a court, not a “fact” question to be decided by experts, no matter how credentialed. To be sure, courts must make a determined effort to grasp the relevant history bearing on that legal question. Hilsenrath, 136 F.4th at 491 (“This kind of historical inquiry requires serious work.” (citation omitted)); McDonald v. City of Chicago, 561 U.S. 742, 803 (2010) (Thomas, J., concurring) (noting “[h]istorical analysis can be difficult”). See generally Heller, 554 U.S. at 592–95, 600–03, 605–19. They do so by consulting articles, books, and historical sources and bringing their own independent judgment to bear on them—not by appointing an “expert,” whose “findings” are insulated by clear-error review on appeal.57 _________________

And then, in one last bizarre twist, the Court says that judges aren't allowed to ask experts in history about questions of history, because experts impact "independent judgment."

Brb going to scream.

The key phrase—“an establishment of religion”—was readily understandable to founding-era citizens. See District of Columbia v. Heller, 554 U.S. 570, 576–77 (2008) (relying on a phrase’s “[n]ormal meaning . . . known to ordinary citizens in the founding generation”). The reason is simple. At the time, establishments were “a familiar institution.” McConnell, Establishment, supra note 12, at 2107.13 Someone on the streets of 1789 Boston, reading that phrase, would have instantly thought of the Church of England, the colonial established churches, or the current state establishments—in other words, a polity’s official church or religion. Ibid.

Although the colonial establishments became more tolerant of dissenters as independence approached, their essence remained unchanged. The original state constitutions reflect as much. Far from rejecting establishments, many states preserved the core components of their establishments, such as public financial support for the official church, regulation of religious institutions, and religious qualifications for civic participation.36 Most explicit was South Carolina, whose 1778 Constitution declared that “the Christian Protestant religion” was “the established religion,” requiring religious societies to subscribe to enumerated articles of faith to receive legal recognition. S.C. Const. of 1778, art. XXXVIII, reprinted in Poore, State Constitutions, supra note 36, at 1626.

The Fifth Circuit flatly states that when the First Amendment says Congress may not create an "establishment of religion," it means the Church of England. They then argue the Founders intended states to have their own churches unaffected by the First Amendment (!!!).

Why is this video not censored? YouTube video by Etymology Nerd

Oh! On Algospeak! you probably don't need to use it now. If you ever actually did. youtube.com/shorts/Z5fvm...

it's very nice to think that we can fix the country and save democracy by playing nice and hoping our fascist pedophile president feels bad and starts playing by the rules but unfortunately that doesn't appear to be happening

A screenshot of clearsky, showing that the bluesky staff member, who goes by Why, has subscribed to only one blocking list, one made by a known troll that searches for users that are against ai and add them to this moderation list. Why @why.bsky.team Blocking 310,708 total users via lists @antiantiai.bsky.social Anti "Anti-AI" Labeler 3 hours ago AI Haters Subscribes to anti-AI blocklists or otherwise seems poorly disposed for productive conversations.

Lmao when a staff member subscribes to a block list made by a troll account and ends up blocking 310,700+ users, all because the majority of the users of bluesky hold the rational opinion that ai is total shit and shouldn't be used for any reason.
Ai is ruining the site, and the world.
Fuck ai.

i wasn't using it but it is such unserious site things that access to the Quiet Posters feed, which is part of the functionality the site has advertised on official accounts just broke for 300k people because it runs out of the tech advisor's personal account and he's a thin skinned AI baby