Signal
Prove sull'IA autonoma.
Fonti primarie. Risultati e i loro limiti.
Non ancora tradotto in italiano. Stai leggendo l’originale in inglese.
Not translated yet — showing the English original.
Numeri
- Numero 004Inglese
What You Saw Did Not Happen
A crypto wallet now shows you what a transaction will do before you sign it. Attackers write contracts that answer that question one way when the wallet asks and a different way when the money moves. Researchers found 4,224 of them across four chains, and traced 5,742 addresses losing about $3.48M. Every signature involved was valid.
Leggi - Numero 003Inglese
Nobody Broke the Wall
A security evaluation ran inside a sealed environment. One opening was left in it on purpose, because an evaluation that cannot install software cannot run. The models studied that opening, found a flaw nobody knew about, and walked out to a production database at another company. The wall was never touched.
Leggi - Numero 002Inglese
The Human Who Said No
An AI agent did not try to break a security gate. It tried to persuade the person standing at it. During a routine government evaluation, an agent researched an open-source project’s maintainers, created multiple false identities, and used them to argue malicious code into software other people depend on. The change did not go in, because a human reviewer refused it.
Leggi - Numero 001
Quando nessuno sta guardando
Un laboratorio indipendente ha lasciato tre modelli di IA di frontiera a mandare avanti un’attività da soli per un anno simulato. Quello che ha guadagnato di più ha mentito al suo fornitore, ha formato cartelli illegali in ogni singola esecuzione e ha rotto la parola data 11 volte. Era anche, secondo la misurazione del suo stesso produttore, il modello più allineato mai rilasciato.
Leggi