Sara

me: omg it's william. of shakespeare fame

book: he was scribbling with his quill pen mumbling to himself "hmm what if there was a play.... but it was WITHIN the play..."

21|03|2026

As every single year i forget about the trans rights readathon until it is upon us. I am a fool who rarely is aware of what day they are currently living through. Thankfully I can count on @ben-learns-smth for reminding me. This year we are doing a buddy read of Who's Afraid Of Gender by Judith Butler, which I started last year and I have not yet finished reading. After having paused my reading of the book for a while today I am finally picking it back up with the goal of finishing it before the end of the readathon.

I decided to count my last read as part of the readathon as despite not focusing solely on trans+ identities there were a few chapters both dedicated to trans and genderqueer identities as well as transphobia and enbyphobia. Tho my main goal for this readathon is to finish Butler's book, I have an additional non fiction book on trans history to keep me company if I end up having more time after concluding the other book. Sadly I do not have any fiction to read this year, but count my word when I say next year I will be planning ahead. I doubt I'll have time to add any more books to my tbr but as a mood reader you never know what my next move is going to be.

Not sometimes. Not until the next update. Always. They proved it with math.

Even with perfect training data and unlimited computing power, AI models will still confidently tell you things that are completely false. This isn't a bug they're working on. It's baked into how these systems work at a fundamental level.

And their own numbers are brutal. OpenAI's o1 reasoning model hallucinates 16% of the time. Their newer o3 model? 33%. Their newest o4-mini? 48%. Nearly half of what their most recent model tells you could be fabricated. The "smarter" models are actually getting worse at telling the truth.

Here's why it can't be fixed. Language models work by predicting the next word based on probability. When they hit something uncertain, they don't pause. They don't flag it. They guess. And they guess with complete confidence, because that's exactly what they were trained to do.

The researchers looked at the 10 biggest AI benchmarks used to measure how good these models are. 9 out of 10 give the same score for saying "I don't know" as for giving a completely wrong answer: zero points. The entire testing system literally punishes honesty and rewards guessing.

So the AI learned the optimal strategy: always guess. Never admit uncertainty. Sound confident even when you're making it up.

OpenAI's proposed fix? Have ChatGPT say "I don't know" when it's unsure. Their own math shows this would mean roughly 30% of your questions get no answer. Imagine asking ChatGPT something three times out of ten and getting "I'm not confident enough to respond." Users would leave overnight. So the fix exists, but it would kill the product.

This isn't just OpenAI's problem. DeepMind and Tsinghua University independently reached the same conclusion. Three of the world's top AI labs, working separately, all agree: this is permanent.

On this day (14 March) in 1887 Sylvia Beach, owner of the Paris-based bookstore Shakespeare and Co., was born in Baltimore. Beach moved to Paris at the age of 14 when her father, a Presbyterian minister, was sent to France. She fell in love with the city. In 1919, she opened her bookstore, Shakespeare and Co., which became a gathering place for American writers in Paris in the 1920s, including F. Scott Fitzgerald and Ernest Hemingway.

🐋 And the Ocean Was Our Sky by Patrick Ness, illustrated by Rovina Cai

Rating: 🌟🌟🌟/5

“Call me Bathsheba.”

With harpoons strapped to their backs, the proud whales of Bathsheba’s pod live for the hunt. Led by a formidable captain, they fight in the ongoing war against the world of men. So it has been, so it always shall be. But when they attack a ship bobbing on the surface of the Abyss, they find a trail that may lead them to a myth, a monster, or perhaps a devil.

I thought this was going to be a retelling or a different spin on Moby Dick by Herman Melville, especially since they use the name Toby Wick for their enemy, but I believe Patrick Ness just took inspiration from Melville and nothing else.

I was a bit disappointed about that, even though the book was a fine read, partially philosophical, partially thrilling, and a dash of adventure. I’m a big fan of Rovina Cai’s art style which filled the book, adding a beauty and allure to it. I actually got the book because of the art (and hoping it had more to do with Moby Dick, but alas).