The market size change in the 'Machine Learning' segment of the artificial intelligence market worldwide was modeled to amount to 203.79 percent in 2025. Between 2021 and 2025, the market size change rose by 154.04 percentage points, though the increase followed an uneven trajectory rather than a consistent upward trend. The market size change is expected to drop by 192.83 percentage points between 2025 and 2032, showing a continuous downward movement throughout the period.Further information about the methodology, more market segments, and metrics can be found on the dedicated Market Insights page on Machine Learning.

Complete dataset included in the full report. Detailed tables, regional splits, forecasts, and methodologies are available with purchase.

In 2025, the market size change in the 'Natural Language Processing' segment of the artificial intelligence market worldwide was modeled to be 87.32 percent. Between 2021 and 2025, the figure dropped by 24.04 percentage points, though the decline followed an uneven course rather than a steady trajectory. The forecast shows the market size change will steadily decline by 64.36 percentage points from 2025 to 2032.Further information about the methodology, more market segments, and metrics can be found on the dedicated Market Insights page on Natural Language Processing.

The artificial intelligence market is estimated to grow from $273.6 billion currently to $5,267 billion by 2035, at a CAGR of 30.84% during the forecast period.

Epoch’s AI models dataset is a collection of machine learning models useful for research about trends in the history and future of artificial intelligence.

Large-scale models are models that were trained with over 10^23 FLOP of compute, which is a static threshold that is used in some AI regulatory frameworks.

By 2035, the Artificial Intelligence (AI) Market is projected to reach USD 3520.9 Bn, up from USD 244.3 Bn in 2025, at a CAGR of 30.6%.

There are the underlying data for our report "Artificial Intelligence: How knowledge is created, transferred, and used", published 2018. Data can be used to construct the graphs used in the report.

The market size in the 'Autonomous & Sensor Technology' segment of the artificial intelligence market worldwide was modeled to amount to 57.3 billion U.S. dollars in 2025. Between 2020 and 2025, the market size rose by 24.52 billion U.S. dollars, though the increase followed an uneven trajectory rather than a consistent upward trend. The market size will steadily rise by 46.04 billion U.S. dollars over the period from 2025 to 2032, reflecting a clear upward trend.Further information about the methodology, more market segments, and metrics can be found on the dedicated Market Insights page on Autonomous & Sensor Technology.

Artificial Intelligence (AI) market size was valued at USD 629.97 Bn in 2026 and is forecast to reach USD 3,796.05 Bn by 2035, at a 21.3% CAGR from 2027 to 2035 .

According to a global survey conducted among 52 participating institutions in 2025, there were more participants who viewed AI's impact on education as an opportunity in every region than as a challenge. However, in the USA and Canada, 43 percent of faculty thought AI could be a challenge in education, higher than in every other region. On the other hand, Latin America had the highest share of respondents viewing AI's impact on education as an opportunity, at 78 percent.

Epoch’s AI models dataset is a collection of machine learning models useful for research about trends in the history and future of artificial intelligence.

The database covers AI models, especially models that are notable for advancing the state of the art, or a large impact on the world or the history of the field.

The data contain several subsets, which can be viewed using the interactive visualization. Notable AI models are those that meet our notability criteria. Frontier models are models that were in the top 5 of training compute as of the time of their release. Large-scale models are models trained with at least 10^23 floating-point operations. All models shows every model in the dataset, including models that do not qualify for the above categories, such as models found during our research on algorithmic progress, AI for biology, or other investigations.

The dataset was originally created for the report “Compute Trends Across Three Eras of Machine Learning” and has continually grown and expanded since then.

Where possible, we collect details such as training compute directly from publications. Otherwise, we estimate details from information such as model architecture and training data, or training hardware and duration.

The dataset is available on our website as a visualization or table, and is available for download in CSV format, updated daily.

The market size change in the 'Autonomous & Sensor Technology' segment of the artificial intelligence market worldwide was modeled to amount to 91.58 percent in 2025. Between 2021 and 2025, the market size change rose by 39.83 percentage points, though the increase followed an uneven trajectory rather than a consistent upward trend. The market size change is expected to drop by 85.3 percentage points between 2025 and 2032, showing a continuous downward movement throughout the period.Further information about the methodology, more market segments, and metrics can be found on the dedicated Market Insights page on Autonomous & Sensor Technology.

Complete dataset included in the full report. Detailed tables, regional splits, forecasts, and methodologies are available with purchase.

The market size in the 'Natural Language Processing' segment of the artificial intelligence market worldwide was modeled to be 126.05 billion U.S. dollars in 2025. Between 2020 and 2025, the market size rose by 104.2 billion U.S. dollars, though the increase followed an uneven trajectory rather than a consistent upward trend. The market size will steadily rise by 742.68 billion U.S. dollars over the period from 2025 to 2032, reflecting a clear upward trend.Further information about the methodology, more market segments, and metrics can be found on the dedicated Market Insights page on Natural Language Processing.

AI statistics for B2B technology marketers — every figure traceable to the primary vendor or research-firm report. Coverage spans enterprise AI adoption, generative AI risk, AI-augmented threat detection, and AI infrastructure spending, reviewed by Nick Cavalancia.

This dataset was created to measure the impact of artificial intelligence on society. The data was collected in Turkey for a Public Opinion Research course, using snowball and convenience sampling methods.

The dataset consists of 205 entries and 20 variables.

The last three variables were designed to assess the participants' knowledge level of artificial intelligence. The intention was to create a new variable by scoring correct answers as '1' and incorrect answers as '0', and then to analyze the relationship between this new score and the other variables.

The majority of the variables were conceived as Likert scale questions. The task of assigning numerical values has been left to the person who will analyze the data.

The dataset is composed entirely of real data.

***Here are the detailed descriptions for each column in the dataset.

• ID: A unique identifier for each survey respondent.

• What is your age range?: The age group of the participant.

• What is your gender?: The gender of the participant.

• What is your education level?: The highest education level completed by the participant.

• What is your employment status?: The current employment status of the participant.

• What is your occupation? (optional): The self-reported occupation of the participant. This field is optional.

• How often do you use technological devices?: The participant's self-reported frequency of daily technology use.

• How much knowledge do you have about artificial intelligence (AI) technologies?: The participant's self-assessed level of knowledge about AI technologies.

• Do you generally trust artificial intelligence (AI)?: The participant's general level of trust in artificial intelligence.

• Do you think artificial intelligence (AI) will be generally beneficial or harmful to humanity?: The participant's opinion on whether AI will ultimately be beneficial or harmful to humanity.

• Please rate how actively you use AI-powered products in your daily life on a scale from 1 to 5.: A numerical rating provided by the participant, from 1 (not active at all) to 5 (very active), on their usage of AI products.

• Would you like to use more AI products in the future?: The participant's stated desire to use more AI-powered products in the future.

• I think artificial intelligence (AI) could threaten individual freedoms.: The participant's level of agreement with the stat- ement that AI could pose a threat to individual freedoms.

• Could artificial intelligence (AI) completely eliminate some professions?: The participant's opinion on the possibility of AI completely replacing some job professions.

• Do you think your own job could be affected by artificial intelligence (AI)?: The participant's belief on whether their own job could be impacted by AI.

• Do you believe that artificial intelligence (AI) should be limited by ethical rules?: The participant's level of agreement with the statement that AI development and use should be constrained by ethical rules.

• Could artificial intelligence (AI) one day become conscious like humans?: The participant's opinion on the possibility of AI achieving human-like consciousness.

• Which of the following do you think is NOT an artificial intelligence (AI) application?: A multiple-choice question to test the participant's knowledge of what constitutes an AI application.

• Which of the following is a machine learning algorithm used in the field of artificial intelligence?: A multiple-choice question to test the participant's knowledge of common machine learning algorithms.

• The artificial intelligence application called 'ChatGPT' is an example of which type of AI system?: A multiple-choice question to test the participant's knowledge about the classification of specific AI systems like ChatGPT.***

Epoch AI’s Biology AI models dataset is a collection of machine learning models trained on biological data, and key information about their training. This dataset is useful for research about trends in the application of artificial intelligence to biology.

The dataset focuses on biological ML models: those that are trained on biological data, including biological sequences, molecular structures or data about molecular properties, among others.

This dataset captures insights into the use of Artificial Intelligence (AI) among 535 students in Indonesian higher education, focusing on their expectations, challenges, attitudes, perceptions, and motivations regarding AI-based learning tools. Collected through a structured survey, the dataset includes demographic variables such as university type, field of study, and educational level, along with students' self-reported experiences with AI in academic settings. The dataset serves as a valuable resource for understanding AI adoption trends in higher education, identifying barriers to AI integration, and evaluating its impact on student engagement and learning outcomes. It enables comparative analysis across different academic disciplines and institutional contexts, offering opportunities for policymakers and educators to design AI-informed curricula. Additionally, this dataset is structured for reproducibility and reuse, allowing researchers to extend its findings, apply alternative analytical approaches, and conduct cross-regional or longitudinal studies on AI integration in higher education.

Public Act No. 23-16, Section 1, effective July 1, 2023, directs the Department of Administrative Services to conduct an annual inventory of all systems that employ artificial intelligence and are used by any state agency.

The public act defines artificial intelligence as (A) an artificial system that (i) performs tasks under varying and unpredictable circumstances without significant human oversight or can learn from experience and improve such performance when exposed to data sets, (ii) is developed in any context, including, but not limited to, software or physical hardware, and solves tasks requiring human-like perception, cognition, planning, learning, communication or physical action, or (iii) is designed to (I) think or act like a human, including, but not limited to, a cognitive architecture or neural network, or (II) act rationally, including, but not limited to, an intelligent software agent or embodied robot that achieves goals using perception, planning, reasoning, learning, communication, decision-making or action, or (B) a set of techniques, including, but not limited to, machine learning, that is designed to approximate a cognitive task.

Integrated Systems Health Management includes as key elements fault detection, fault diagnostics, and failure prognostics. Whereas fault detection and diagnostics have been the subject of considerable emphasis in the Artificial Intelligence (AI) community in the past, prognostics has not enjoyed the same attention. The reason for this lack of attention is in part because prognostics as a discipline has only recently been recognized as a game-changing technology that can push the boundary of systems health management. This paper provides a survey of AI techniques applied to prognostics. The paper is an update to our previously published survey of data-driven prognostics.