Sarah Rajtmajer

@bwyble.bsky.social a propos of our conversation today… 🤯

Time machine

👿 "ChatGPT “encouraged him to continue his podcast because it was creating ‘haters,’ which meant monetization for Dadig,” the DOJ alleges. He also claimed that ChatGPT told him that “people are literally organizing around your name, good or bad, which is the definition of relevance,..."

The ❄️ ❄️ ❄️ is here. Upside: ⛷️ season

Even the the damn twitter card for this Nature Scientific Reports is clearly AI Slop.

Screenshot of a tweet by Melanie D’Arrigo on X.com that reads: “It’s literally the whole reason why we celebrate Thanksgiving.” Beneath it is a Homeland Security tweet saying: “This Thanksgiving, there is no room at the table for invaders.”

The irony is rich.

pura vida ✨

Although most of my photos right now are of State College autumn, here's Happy Valley from the sky this August. It was awe-inspiring to get this view.

So, come join us here! I'll be recruiting a graduate student to the Empathy & Moral Psychology (EMP) Lab during this cycle, to start in fall 2026.

you know what? good advice.

695 pages of machine language, 80,000 computational cycles, representing 0.01 seconds of sound.
I have always Felt music to be so deeply human. Seeing it represented this way takes on new meaning in the era of AI.

Timeless @tristanperich.bsky.social

The CDC website now disseminates disinformation about vaccines, claiming erroneously that infant vaccines might cause autism, when we know conclusively that they do not. It is difficult to overstate just how dangerous this is. www.cdc.gov/vaccine-safe...

What AI companions can’t give.

"My Boyfriend is AI": A Computational Analysis of Human-AI Companionship in Reddit's AI Community The emergence of AI companion applications has created novel forms of intimate human-AI relationships, yet empirical research on these communities remains limited. We present the first large-scale com...

These examples are from a recent mixed-methods study from MIT (arxiv.org/abs/2509.11391) that also found that "AI companionship rarely begins intentionally: 10.2% developed relationships unintentionally through productivity-focused interactions, while only 6.5% deliberately sought AI companions."

‘Mind-captioning’ AI decodes brain activity to turn thoughts into text A non-invasive imaging technique can translate scenes in your head into sentences. It could help to reveal how the brain interprets the world.

Next in the emerging-tech series of: What could possibly go wrong? 😅
www.nature.com/articles/d41...

@franciscopereira.bsky.social - saw this and thought of you immediately, of course. Reminded me of many wonderful chats in New Jersey.

Exactly.

“Jesus App Encouraged Teen to Kill Parents, Commit Suicide”

We just donated through PASCP - they are also collecting for E.A.T.S.:)

The Securing American Funding and Expertise from Adversarial Research Exploitation (SAFE) Act would deny federal funding to any U.S. scientist who collaborates with anyone “affiliated with a hostile foreign entity,” a category that includes four countries: China, Russia, Iran, and North Korea. The prohibited activities would include joint research, co-authorship on papers, and advising a foreign graduate student or postdoctoral fellow. The language is retroactive, meaning any interactions during the previous 5 years could make a scientist ineligible for future federal funding.

I've been on the road so I'm behind the times—but if you wanted to destroy US science, I can think of no more expedient action.

Blatant unconstitutionality aside, fuck this backwards forwards and sideways.

www.science.org/content/article/u-s-congress-considers-sweeping-ban-chinese-collaborations

YouTube video by More Perfect Union Victims Sue Sam Altman for ChatGPT Encouraging Harm and Suicide

Life imitating art or vice versa? I'm not sure, but "Victims sue Sam Altman for ChatGPT encouraging harm and suicide" is the plot of "The Proving Ground," the latest "Lincoln Lawyer" novel from Bosch author @michaelconnelly.bsky.social

Perfect

Opinion | Did Liberal Feminism Ruin the Workplace?

This coming on the heels of: www.nytimes.com/2025/11/06/o...
Believe it or not, this is the new, more politically-correct headline. The original title was "Did women ruin the workplace?" but that one surprisingly garnered some pushback.
Good for attention, though.

It seems increasingly frequent that my day gets punctuated by notifications from the Economist or NYTimes or other formally reputable source framing urgent truths as “questions”… #unsubscribe.

Imagination is resistance, @ruha9.bsky.social reminds us that, " we need to reject the gospel that says technology is inevitable, that the train has left the station... In an error of automated bullshit, it’s time to become bullshit detectors,"
#MozFest

A Tale of Two Identities: An Ethical Audit of Human and AI-Crafted Personas As LLMs (large language models) are increasingly used to generate synthetic personas particularly in data-limited domains such as health, privacy, and HCI, it becomes necessary to understand how these...

"A Tale of Two Identities: An Ethical Audit of Human and AI-Crafted Personas" is accepted to AAAI-26! @aaai.org
@shomir.bsky.social

We show LLMs demonstrate "algorithmic othering"--e.g., disproportionately foreground racial markers, overproduce culturally coded language.

arxiv.org/abs/2505.07850

This is really impressive. They invented racism.

@vrandecic.bsky.social founder of Wikidata shares the history of Wikipedia and Semantic Web. @iswc-conf.bsky.social

This example from early days a nod to his beautiful HR

Reminder of the power of a kind word.
I gave my talk (a thought piece/passion project). Someone in the audience raised his hand and said it was one of the most inspiring talks he's ever heard.

Such a meaningful compliment, I will never forget it.
When you think something nice, say it out loud.

I missed the memo about the dress code for this flight. #Tokyo

And congrats are in order - Diella is pregnant! www.euronews.com/next/2025/10...
Is it rude to ask who the father is?

Beyond the problems outlined by TechPolicy, below, the anthropomorphism here is out of hand.

Croatian Serbs Urge Prosecutions Over Jasenovac Camp Denial in Parliament The Serbian National Council said right-wingers who used a discussion in parliament to reject facts about crimes committed at the World War II-era Jasenovac camp violated the Croatian criminal code, w...

Milorad Pupovac, always antifascist ✊ he's right, and I hope action is taken against Ustasha apologetics in the Sabor.