Scott Alexander at Astral Codex Ten:
What’s the difference between ‘real’ and ‘roleplaying’?
One possible answer invokes internal reality. Are the AIs conscious? Do they “really” “care” about the things they’re saying? We may never figure this out. Luckily, it has no effect on the world, so we can leave it to the philosophers1.
I find it more fruitful to think about external reality instead, especially in terms of causes and effects.

Does Moltbook have real causes?If an agent posts “I hate my life, my human is making me work on a cryptocurrency site and it’s the most annoying thing ever”, does this correspond to a true state of affairs? Is the agent really working on a cryptocurrency site? Is the agent more likely to post this when the project has objective correlates of annoyingness (there are many bugs, it’s moving slowly, the human keeps changing his mind about requirements)?
Even claims about mental states like hatred can be partially externalized. Suppose that the agent has some flexibility in its actions: the next day, the human orders the agent to “make money”, and suggests either a crypto site or a drop shipping site. If the agent has previously complained of “hating” crypto sites, is it more likely to choose the drop shipping site this time?
More here.
Enjoying the content on 3QD? Help keep us going by donating now.


We think the current evidence is clear. By inference to the best explanation — the same reasoning we use in attributing general intelligence to other people — we are observing AGI of a high degree. Machines such as those envisioned by Turing have arrived. Similar arguments have been made before, and have engendered controversy and push-back. Our argument benefits from substantial advances and extra time. As of early 2026, the case for AGI is considerably more clear-cut.
NEW YORK — A Doberman pinscher named Penny was awarded best in show at the Westminster Kennel Club Dog Show on Tuesday night, the fifth of its breed to take the title of top dog. “You can’t attribute it to one thing, but she’s as great a Doberman as I’ve ever seen,” said her handler, Andy Linton, who previously won best in show 37 years ago and had hoped for a second victory, adding, “I had some goals, and this was one of them.”
I
Daniel Kahneman and Amos Tversky’s joint papers from the 1970s and 1980s have inspired many, including myself. These articles magically turned statistical thinking—previously a niche interest—into a major re-search focus. Kahneman and Tversky revived the concept of heuristics, which had largely been forgotten at the time, and played a pivotal role in bringing psychology to the attention of economics and other social sciences. I was also deeply influenced by Tversky’s seminal work on the foundations of measurement, which inspired my first book on modeling.
On Jan. 28, 2026, President Donald Trump sharply intensified his threats to the Islamic Republic, suggesting that if Tehran did not agree to a set of demands, he could mount an attack “
The adventure story and the historical romance were two genres at which Stevenson excelled, but he was also brilliant at the macabre psychological parable in his novella The Strange Case of Dr. Jekyll and Mr. Hyde (1886), and the supernatural in his short story “Thrawn Janet” (1881). The first of these takes on the very “fortress of identity” (in Jekyll’s words) that has so obsessed us of late but turns it into something timeless. Damrosch tells us that the novella caused a furious argument between Stevenson and his wife, in which she comes off better than he does. When Louis read aloud his first draft, as Fanny’s son Lloyd recalled, “Her praise was constrained; the words seemed to come with difficulty; and then all at once she broke out with criticism. He had missed the point, she said; had missed the allegory; had made it merely a story—a magnificent bit of sensationalism—when it should have been a masterpiece.” Damrosch continues, “Fanny’s point was that Louis had ruined the story by turning it into a mere tale about a secret life. . . . What was needed was not just a character wearing a disguise, but something far more profound: a character struggling with a deeper hidden self that breaks loose and fights for supremacy.” Louis resisted, then came around, went back to work, and gave her the masterpiece she wanted. Thereafter, he jokingly referred to her as “the critic on the hearth.”
Popular wisdom holds we can ‘rewire’ our brains: after a stroke, after trauma, after learning a new skill, even with
Creativity is a trait that AI critics say is likely to remain the preserve of humans for the foreseeable future. But a large-scale study finds that leading generative language models can now exceed the average human performance on linguistic creativity tests.
So I was thinking about the old logic problem/koan
Dr Johnson never filmed a “spicy books with cartoon covers” vlog. But Jack Edwards cannot quite deny being the most important literary critic in the world. In commercial terms, he certainly is. A nod from him fills bathtubs, train carriages and public parks with copies of a book he likes. Booksellers buy and arrange their stock to his taste. And he is not confined to new releases. When he dug up an obscure Dostoevsky (White Nights), his positive review moved it from cellars to shop windows instantaneously. I first met him for this interview around the time of the 2024 International Booker Prize. He had been asked to host the ceremony – and to livestream it. I watched him cruise up the red carpet, encircled by cameras and attendants.
The backstory: a few months ago, Anthropic released Claude Code, an exceptionally productive programming agent. A few weeks ago, a user modified it into Clawdbot, a generalized lobster-themed AI personal assistant. It’s free, open-source, and “empowered” in the corporate sense – the designer