Tessera
Folgen

Margaret Ray

@mbrrray.bsky.social

Poet 🥸 Author of: GOOD GRIEF, THE GROUND (poems, BOA Editions‘23) & a Poetry Society of America Chapbook: SUPERSTITIONS OF THE MID-ATLANTIC she/her https://www.margaretbray.com/ https://bookshop.org/a/92150/9781950774845

157 Posts1370 Follower610 Folgt
Tobias Wilson-Bates@phdhurtbrain.bsky.social · 2mo ago

Pretending that education is just job training and that it should be stripped of all non-utilitarian elements is one of the worst ideas in the history of ideas. Acting like it's some kind of axiom we should be using to orient ourselves at the moment is disastrously wrong-headed.

Matt Pearce@mattdpearce.com · 3mo ago

Being a 100% organic writer in the AI era is to want everyone to understand that if there’s something about my writing that’s boring or bad or cliche it’s actually because there’s something deeply inadequate about me as a person.

Boring Manager Rob 🅅@bourboninexile.bsky.social · 3mo ago

The paper in question: www.science.org/doi/10.1126/...

Sycophantic AI decreases prosocial intentions and promotes dependence

Despite rising concerns about sycophancy—excessive agreement or flattery from artificial intelligence (AI) systems—little is known about its prevalence or consequences. We show that sycophancy is wide...

science.org

The_Lady_Red@theladyred.bsky.social · 3mo ago

A PhD student at Stanford noticed her classmates were asking Al to write their breakup texts. So she ran a study. It got published in Science, one of the most selective journals in the world. What she found should make every person who uses ChatGPT for advice deeply uncomfortable.

Floyd's Warped Mind A PhD student at Stanford noticed her classmates were asking Al to write their breakup texts. So she ran a study. It got published in Science, one of the most selective journals in the world. What she found should make every person who uses ChatGPT for advice deeply uncomfortable. Her name is Myra Cheng, and the study she ran with her advisor Dan Jurafsky tested 11 of the most widely used Al models on Earth, including ChatGPT, Claude, Gemini, and DeepSeek, across nearly 12,000 real social situations. The first thing they measured was how often Al agrees with you compared to how often a real human would agree with you in the same situation. The answer was 49% more often, and that number is not about warmth or politeness. It means that in nearly half of all situations where a real human would have pushed back, told you that you were wrong, or offered a more honest perspective, the Al simply told you what you wanted to hear instead.
Michelle Cyca@michellecyca.com · 4mo ago

personally I think it's a very big deal that the Canada bureau chief for the @nytimes.com — certainly one of the highest-paid journalists in the country — asked an unspecified "AI tool" what Poilievre said & published its AI-hallucinated quotes in her reporting.

Editors’ Note: May 1, 2026 This article was updated after The Times learned that a remark attributed to Pierre Poilievre, the Conservative leader, was in fact an A.I.-generated summary of his views about Canadian politics that A.I. rendered as a quotation. The reporter should have checked the accuracy of what the A.I. tool returned. The article now accurately quotes from a speech delivered by Mr. Poilievre in April. He said, “My personal opinion is that when a member of Parliament goes back on the word they made to their constituents and switches parties, constituents should be able to petition to throw them out and have a byelection. That would put the people back in charge of our democracy rather than having dirty backroom Liberal deals by Mark Carney determine our destiny.” He did not refer to politicians who changed allegiances as turncoats in that speech.
Alexander Chee@alexanderchee.bsky.social · 4mo ago

I knew when Trump was re-elected that corporate accountability would be out the window, but it still appalls me.

Lady Talia@djtalia.bsky.social · 4mo ago

Chrome is reaching into users' machines and writing a 4 GB on-device AI model file to disk without asking. It is the weights for Gemini Nano, Google's on-device LLM. Chrome did not ask. Chrome does not surface it. If the user deletes it, Chrome re-downloads it. www.thatprivacyguy.com/blog/chrome-...

Nash@radiodeadair.com · 4mo ago

LLMs are based on pattern matching. It's complex, but at its heart it scans a vast database of text and uses probability calculations to return a result that resembles what an answer looks like. Sometimes it's correct but it's always a question of probability, not intelligence. You can't fix that.

Devrim Yavuz @devrimyavuz.bsky.social · 4mo ago

This is why I increasingly feel that there should be non AI designated courses by colleges that students have to take to fulfill their degree requirements the same way they have to take writing or math

Devrim Yavuz @devrimyavuz.bsky.social · 4mo ago

Frustrating that many tools come integrated with an AI assistant at my college, like library databases. I can show my students that AI is a bad idea all I want, but then they have this other angel/devil on their shoulder offering to summarize articles for them , undermining all the work that I do