Claude Fable 5 doesn’t truly understand. And here is a beautiful proof:
The Beninatto-Trombetti test is a translation test for professional translators. It measures the ability to infer context, revise the surface form, and generalize beyond literal mapping.
For example, the correct translation of:
“Solo 3 parole: non sei solo”
is not:
“Just 3 words: you are not alone”
but:
“Just 4 words: you are not alone.”
An LLM that understands the sentence must also update the meta-linguistic claim inside the sentence.
Claude Fable 5 is arguably the most advanced LLM currently available. And yet it still fails this simple test.
LLMs are extraordinary machines for recombining existing knowledge. But they don’t truly understand.
We are still far from AGI.
AI-powered computer worm, a self-replicating agent that reasons its way through a network instead of carrying a fixed exploit list. It steals compute from compromised GPU machines to run its own open-weight LLM, then uses weaker machines as relays for reach. In trials on a corporate testbed, it identified vulnerabilities, exploited systems, and launched replicas across Linux, Windows, and IoT targets. Every new infection can add more infrastructure while costing the attacker almost nothing. Patching one flaw no longer ends the threat, because the worm can operationalise fresh advisories, generate new attack logic, and keep adapting without a human operator. It is not a WannaCry-style worm with one baked exploit and one baked ransomware payload. It can adapt across many vulnerability classes it can discover and operationalise https://t.co/nSupd1h0BG
“Sunday is the new Sabbath” was basically the slogan with which, in the 4th century, with a fair share of anti-Semitism, Christianity codified Sunday as a day of rest in civil and ecclesiastical law.
https://t.co/pN8kpEyxsW
What looks like harmless restoration, such as colorizing old photographs, can quietly replace historical evidence with synthetic memory.
Letting AI reimagine the past is probably a bad idea.
https://t.co/EShDp7uqP6
“You know, they take a pill to try and stay up to write an essay all night and end up cleaning the bathroom.”
Smart drugs promise focus and intelligence, but the evidence points to a less glamorous answer: fatigue is not a problem we can outsmart forever.
https://t.co/QVqH3KQCFL
“I believe that an orderly universe, ... in which everything has an explanation even if we still have a long way to go before we find it, is a more beautiful, more wonderful place than a universe tricked out with capricious, ad hoc magic.”
https://t.co/zl4UwLGOYh
“I believe that an orderly universe, ... in which everything has an explanation even if we still have a long way to go before we find it, is a more beautiful, more wonderful place than a universe tricked out with capricious, ad hoc magic.”
https://t.co/zl4UwLGOYh
Active democracy, economic equality, considerable racial tolerance, and even medical coverage — are characteristic of contemporary democracies and were present aboard pirate ships long before American Independence and the French Revolution.
https://t.co/Af2Ig68HtM
"Paper Factory"
Multi-agent workflow for producing social science papers from
@natewilmers at MIT
@pengzell at University College London
You can read from a gallery of the output papers to see the quality yourself--I honestly found it a little unsettling.
Or even if you don't run this workflow, just reading the step-by-step txt instructions and thinking about how you'll advise your PhD students in the future seems a worthwhile use of time, for me at least.
Links to github, paper, and gallery of outputs in the quoted post from Nate below:
It’s when we manage to lose the sensory pride of assuming the world is exactly as our species perceives it that the world as others know it can begin to become real.
https://t.co/R7RmNN3Z2u
It’s when we manage to lose the sensory pride of assuming the world is exactly as our species perceives it that the world as others know it can begin to become real.
https://t.co/R7RmNN3Z2u