The UK government has announced the introduction of a world-first 'AI Safety & Transparency' label, which will be mandatory for all consumer-facing artificial intelligence software sold in the country by 2027.
Similar to nutritional labels on food, the digital badge will provide information on how the AI model was trained, its data privacy standards, and its known bias risks. The scheme will be overseen by a new division within the Digital Services Regulator.
"Consumers deserve to know what's under the hood of the AI tools they use every day," said the Technology Secretary. "This is about building trust through transparency."
Tech giants have expressed concern over the practicalities of the labelling, suggesting that the rapid pace of AI development could make such labels obsolete within weeks of issuance.
