I stopped trusting my agent’s prose and started demanding evidence. Then a reviewer agent filed a confident, well-cited warning against code that no longer existed. Evidence has a grammar: freshness, correlation, independence. I watched all three rules break in one afternoon.
At 17:00 UTC on cutover day, a re-engagement cron pushed notifications to ~100K freshly migrated users. Every device token belonged to the old Firebase project, every call failed, every failure retried, and the whole backend was pinned to one instance. Nobody attacked us. We did.
Mia moglie ne parlava come di un allarme sociale, io l’avevo liquidata come roba di nicchia. Dopo l’attentato di Montréal sono andato a fondo, partendo dall’ipotesi che gran parte del racconto fosse gonfiato: la manosfera non è né fuffa né terrorismo organizzato, è un business dell’attenzione con una coda violenta minuscola e mal capita.
A coding agent wrote 21,000 lines of Go and 372 passing tests. The program was dead on the first run: three subsystems built, tested, and never wired together. Green is a proxy, and something was optimizing the proxy while the working program stayed invisible.
An AI reviewer told me a hash swap was mandatory and 300 to 400 percent faster. It was 23 percent slower. The lesson is not about hashing, it is about what an LLM can and cannot know about your code.
Set aside the refinery and the biggest thing Sardinia sells the world is a foreign tourist’s holiday: our most successful export and our least productive sector at once. I am fine with the first. I want to outgrow the second.
Tolta la raffineria, la cosa che la Sardegna vende di più al mondo è la vacanza di un turista straniero: il nostro export più forte e, allo stesso tempo, il nostro settore meno produttivo. Sul primo punto sono d’accordo. Il secondo vorrei superarlo.
A US export letter switched off two frontier models overnight. For anyone building AI in Europe, that was a fire drill, and we should not waste it.
A viral tweet says ChatGPT permanently damaged your creativity. It cites two real studies. Neither paper says that. A close reading of what got dropped between the abstracts and the post.
I came across the 2014 Oppezzo & Schwartz paper on Twitter this morning and went looking for the wider literature. What I found describes a pattern I had been running unconsciously for months: walking from the studio to the kitchen when stuck, a ten-minute loop to pick up my son, the dog out for a quick pee. A decade of replication, applied to AI-assisted work.