Open weights:
Open weights are the publicly released learned parameters of a trained artificial intelligence model, principally its weights and biases. In an artificial neural network, weights are numerical values that determine how strongly inputs contribute to the network’s calculations. Publishing these values allows others to download and use the model, while permission to modify, fine-tune or redistribute it depends on its license. The term is commonly applied to large language models, diffusion models and other generative artificial intelligence.[1][2] Open weights is contrasted to open-source artificial intelligence, which considers the public release of not only parameters but the model’s source code, training data, evaluation results, intermediate checkpoints and technical documentation. [Wikipedia]
The page carries a warning that some of the text may be AI-generated, which is prohibited in Wikipedia articles, and not a practice I condone. Noted in “Anthropic reveals $80 billion loss, surging revenue as it prepares to go public,” Gerrit De Vynck and Miriam Waldvogel, WaPo:
But not everyone is convinced. Some financial analysts have questioned the companies’ abilities to ever become profitable, given the huge costs involved in training AI and the close competition from open-weights AI models, which are nearly as capable as the high-end AI from Anthropic and OpenAI but are much cheaper to operate.
The analog to open source.
Or is it more appropriate to suggest the digital to open source?
Never mind me, I’m a bit short of sleep. Nudge nudge, wink wink.
