Definition
Data poisoning
An attack where someone slips corrupted, altered, or mislabeled examples into a model's training data to make it behave badly or carry a hidden backdoor.
Learn the language of AI-native founders
Every week: the new terms, tools, and workflows founders are using, straight from the podcast.
Related terms in AI Fundamentals
Contrastive learningA training technique where a model learns by comparing similar and dissimilar examples, figuring out what makes things alike or different.Data augmentationExpanding a training dataset by creating modified copies of existing data, like flipped images or reworded sentences, instead of collecting new data.Data Collection (Local AI)A curated subset of a user's files — PDFs, documents, images, voice memos — that is analyzed once by a local AI system and made queryable without any cloud access. The unit of organization for privacy-first AI interaction with personal or proprietary data.Data labelingTagging documents so a model learns what matters in them.Dead Internet TheoryThe concept that AI-generated content is increasingly training on other AI-generated content, creating a feedback loop of degraded and homogenized information. Kaelan references it as the backdrop for why original, experience-grounded content increasingly wins.Decision treeA model that makes predictions by asking a series of yes-or-no questions about the data, branching at each step until it reaches an answer.Deep learningMachine learning that uses neural networks with many layers to learn complex patterns directly from raw data.DeepfakeAI-generated video, image, or audio convincing enough to impersonate a real person.