Scale AI
In plain wordsA company that pays large numbers of people to label examples, write model answers and grade chatbot replies, producing the data AI developers train and test on. Meta bought nearly half of it in 2025.
An American company that sells human-generated training data and model testing to AI developers, governments and businesses. It began with data labelingdata labeling, n.: The work of attaching correct answers to examples — often done by large paid workforces, sometimes by experts — so that AI systems can learn from them or be tested against them. of camera and lidar recordings for self-driving carsautonomous vehicle, n.: A vehicle, usually a car, truck or shuttle, that can drive itself, sensing the road, deciding what to do and steering, braking and accelerating with no human driving., then became a leading supplier of the human-written demonstrations and preference rankings used in supervised fine-tuningsupervised fine-tuning, n.: Teaching a pretrained AI model by showing it many examples of good answers to requests, so it learns to imitate them — like learning a job by studying worked examples. and reinforcement learning from human feedbackreinforcement learning from human feedback, n.: A way of training AI using people's judgments: humans compare pairs of answers, a scoring model learns their preferences, and the AI is then trained to produce answers that score well., drawing on large pools of contract workers through platforms such as Remotasks and Outlier; pay and working conditions on those platforms have been criticized and litigated. It also builds evaluations, co-creating Humanity's Last ExamHumanity's Last Exam, prop. n.: A test of 2,500 extremely hard questions with definite answers, written by about a thousand experts worldwide and billed as the last academic exam AI would need; at launch, top models scored under 10%. with the Center for AI SafetyCenter for AI Safety, prop. n.: A San Francisco nonprofit that studies how advanced AI could lead to disaster and how to prevent it. In 2023 it gathered signatures for a one-sentence warning that AI could threaten human extinction., and develops AI systems for the US military. In June 2025 Meta bought a 49 percent non-voting stake and hired its founder; several AI developers, reportedly including Google and OpenAI, then wound down work with it, citing neutrality concerns.
Origin
Founded in 2016 in San Francisco by Alexandr Wang and Lucy Guo through the Y Combinator startup accelerator; Wang left the Massachusetts Institute of Technology to start it. After Wang joined Meta in June 2025, Jason Droege became chief executive; Francis deSouza was named to the post in July 2026.
Related
- data labelingdata labeling, n.: The work of attaching correct answers to examples — often done by large paid workforces, sometimes by experts — so that AI systems can learn from them or be tested against them.
- reinforcement learning from human feedbackreinforcement learning from human feedback, n.: A way of training AI using people's judgments: humans compare pairs of answers, a scoring model learns their preferences, and the AI is then trained to produce answers that score well.
- Humanity's Last ExamHumanity's Last Exam, prop. n.: A test of 2,500 extremely hard questions with definite answers, written by about a thousand experts worldwide and billed as the last academic exam AI would need; at launch, top models scored under 10%.
- Meta Superintelligence LabsMeta Superintelligence Labs, prop. n.: A division Meta created in 2025 to pursue superintelligence, gathering all of the company's AI teams under new leaders recruited in an expensive hiring campaign.
- ghost workghost work, n.: The hidden human labor behind AI: people, often poorly paid and far away, who label data, check outputs and handle cases the software cannot, while the product appears fully automatic.
Mentioned in
Center for AI SafetyCenter for AI Safety, prop. n.: A San Francisco nonprofit that studies how advanced AI could lead to disaster and how to prevent it. In 2023 it gathered signatures for a one-sentence warning that AI could threaten human extinction., Dan HendrycksDan Hendrycks, prop. n.: An American AI-safety researcher who created widely used AI tests such as MMLU and Humanity's Last Exam, and who runs the Center for AI Safety., Mark ZuckerbergMark Zuckerberg, prop. n.: The American co-founder and chief executive of Facebook's parent company Meta, who made it a major AI developer that released its Llama models for anyone to download., Meta Superintelligence LabsMeta Superintelligence Labs, prop. n.: A division Meta created in 2025 to pursue superintelligence, gathering all of the company's AI teams under new leaders recruited in an expensive hiring campaign., Alexandr WangAlexandr Wang, prop. n.: An American entrepreneur who founded Scale AI, a major supplier of human-labeled data for training AI, and became Meta's chief AI officer in 2025., reverse acquihirereverse acquihire, n.: A deal in which a big tech company hires a start-up's founders and top staff and pays to license its technology, without buying the company itself.