Everything Green, Everything Broken
LLMs in Production, the Failures That Raise No Error and the Five Numbers Nobody Counts
Impossible d'ajouter des articles
Désolé, nous ne sommes pas en mesure d'ajouter l'article car votre panier est déjà plein.
Veuillez réessayer plus tard
Veuillez réessayer plus tard
Échec de l’élimination de la liste d'envies.
Veuillez réessayer plus tard
Impossible de suivre le podcast
Impossible de ne plus suivre le podcast
Bénéficiez gratuitement de Standard pendant 30 jours
5,99 €/mois après la période d’essai. Annulation possible à tout moment
Everything Green, Everything Broken
Lu par :
AI Voice A synthetic voice
Acheter pour 14,99 €
-
Lu par :
-
AI Voice A synthetic voice
-
De :
-
Nikolaos Broikos
Ce titre utilise une narration à voix virtuelle
La voix virtuelle est une narration générée par ordinateur pour les livres audio.
A model refused to answer while the answer sat in front of it, complete and correct, and then invented a data quality problem to justify the refusal, in the register of a careful bug report. A filter identified sensitive material perfectly, labelled it, wrote a log line proving it had seen it, and let it through. An upgrade rewrote the format of more than a quarter of responses while every accuracy test reported a clean migration. A cost model was wrong in two directions at once.
Six failures. Zero errors. Several of them producing positive evidence of health.
None of it was hidden. All of it sat in plain text, in fields nobody aggregated, phrased in a way nothing matched, or summarised by a statistic that absorbs exactly that signal. This book is about the class of failure that leaves no trace your monitoring was built to catch, and the five cheap numbers that catch it instead. None of them needs a labelled dataset, a vendor, a budget or anybody's permission.
The good news is the argument. The unpredictable part of your system was always the small part. The provable part was always most of it, and almost nobody tests it, because the presence of a model made an entire subsystem feel unmeasurable when it is not.
For engineers and technical leaders running something real, who suspect the green dashboard is not the whole story and want to know exactly where to look.
adbl_web_anon_alc_button_suppression_t1
Aucun commentaire pour le moment