woke AI

Political label

A pejorative term, used mainly by conservative critics in the United States, for AI systems perceived as embedding progressive or diversity-oriented viewpoints in their outputs — for example by refusing some politically sensitive requests, giving one-sided answers on contested issues or, in a widely cited February 2024 episode, generating historically inaccurate images of people of varied races in response to prompts about historical figures. Critics of the label argue that it lumps together genuine accuracy failures, deliberate safety measures against harmful content and efforts to reduce algorithmic biasalgorithmic bias, n.: Unfairness in software that judges or serves people: it consistently produces worse results for some groups, such as women, older people or ethnic minorities, often because it learned from skewed or incomplete records., and that political neutrality is itself a contested standard.

The senator called the chatbot woke AI after it declined to write the campaign attack ad.

US federal procurementlegal

Under Executive Order 14319, 'Preventing Woke AI in the Federal Government' (July 23, 2025), federal agencies may procure large language modelslarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. only if they comply with two Unbiased AI Principles: truth-seeking (factual accuracy and acknowledgment of uncertainty) and ideological neutrality (no partisan or ideological judgments built into outputs unless a user asks). Office of Management and Budget memorandum M-26-04 (December 11, 2025) implements it through contract terms requiring vendors to supply documentation such as acceptable-use policies and model, system or data cards and to remedy violations, with limited exceptions such as national-security systems; agencies had until March 11, 2026 to update their procurement policies.