Creativity is core to being human. Generative AI—made readily available by powerful large language models (LLMs)—holds promise for humans to be more creative by offering new ideas, or less creative by anchoring on generative AI ideas. We study the causal impact of generative AI ideas on the production of short stories in an online experiment where some writers obtained story ideas from an LLM. We find that access to generative AI ideas causes stories to be evaluated as more creative, better written, and more enjoyable, especially among less creative writers. However, generative AI-enabled stories are more similar to each other than stories by humans alone. These results point to an increase in individual creativity at the risk of losing collective novelty. This dynamic resembles a social dilemma: with generative AI, writers are individually better off, but collectively a narrower scope of novel content is produced. Our results have implications for researchers, policy-makers, and practitioners interested in bolstering creativity. Methods This dataset is based on a pre-registered, two-phase experimental online study. In the first phase of our study, we recruited a group of N=293 participants (“writers”) who are asked to write a short, eight sentence story. Participants are randomly assigned to one of three conditions: Human only, Human with 1 GenAI idea, and Human with 5 GenAI ideas. In our Human only baseline condition, writers are assigned the task with no mention of or access to GenAI. In the two GenAI conditions, we provide writers with the option to call upon a GenAI technology (OpenAI’s GPT-4 model) to provide a three-sentence starting idea to inspire their own story writing. In one of the two GenAI conditions (Human with 5 GenAI ideas), writers can choose to receive up to five GenAI ideas, each providing a possibly different inspiration for their story. After completing their story, writers are asked to self-evaluate their story on novelty, usefulness, and several emotional characteristics. In the second phase, the stories composed by the writers are then evaluated by a separate group of N=600 participants (“evaluators”). Evaluators read six randomly selected stories without being informed about writers being randomly assigned to access GenAI in some conditions (or not). All stories are evaluated by multiple evaluators on novelty, usefulness, and several emotional characteristics. After disclosing to evaluators whether GenAI was used during the creative process, we ask evaluators to rate the extent to which ownership and hypothetical profits should be split between the writer and the AI. Finally, we elicit evaluators’ general views on the extent to which they believe that the use of AI in producing creative output is ethical, how story ownership and hypothetical profits should be shared between AI creators and human creators, and how AI should be credited in the involvement of the creative output. The data was collected on the online study platform Prolific. The data was then cleaned, processed and analyzed with Stata. For the Writer Study, of the 500 participants who began the study, 169 exited the study prior to giving consent, 22 were dropped for not giving consent, and 13 dropped out prior to completing the study. Three participants in the Human only condition admitted to using GenAI during their story writing exercise and—as per our pre-registration—they were therefore dropped from the analysis, resulting in a total number of writers and stories of 293. For the Evaluator Study, each evaluator was shown 6 stories (2 stories from each topic). The evaluations associated with the writers who did not complete the writer study and those in the Human only condition who acknowledged using AI to complete the story were dropped. Thus, there are a total of 3,519 evaluations of 293 stories made by 600 evaluators. Four evaluations remained for five evaluators, five evaluations remained for 71, and all six remained for 524 evaluators.

Objective: Biomechanical Machine Learning (ML) models, particularly deep-learning models, demonstrate the best performance when trained using extensive datasets. However, biomechanical data are frequently limited due to diverse challenges. Effective methods for augmenting data in developing ML models, specifically in the human posture domain, are scarce. Therefore, this study explored the feasibility of leveraging generative artificial intelligence (AI) to produce realistic synthetic posture data by utilizing three-dimensional posture data.Methods: Data were collected from 338 subjects through surface topography. A Variational Autoencoder (VAE) architecture was employed to generate and evaluate synthetic posture data, examining its distinguishability from real data by domain experts, ML classifiers, and Statistical Parametric Mapping (SPM). The benefits of incorporating augmented posture data into the learning process were exemplified by a deep autoencoder (AE) for automated feature representation.Results: Our findings highlight the challenge of differentiating synthetic data from real data for both experts and ML classifiers, underscoring the quality of synthetic data. This observation was also confirmed by SPM. By integrating synthetic data into AE training, the reconstruction error can be reduced compared to using only real data samples. Moreover, this study demonstrates the potential for reduced latent dimensions, while maintaining a reconstruction accuracy comparable to AEs trained exclusively on real data samples.Conclusion: This study emphasizes the prospects of harnessing generative AI to enhance ML tasks in the biomechanics domain.

Generative AI made the most significant difference in automation potential for those professions needing master’s degrees or higher levels of education. The lack of increase in automation for those with high school diplomas or lower-level education is likely because those with education of that level work highly physical and irregular jobs, activities that are difficult to automate for the digital generative AI process. Generative AI changes the automation trend Before the arrival of powerful new generative AI programs such as ChatGPT or Google’s Gemini, it was generally held that automation was going to hit the blue-collar side of the workforce harder than the white-collar. With generative AI, this has been thrown into an upheaval, with some estimates suggesting that education and workforce training, for example, could be automated at nearly * times the rate before generative AI. Offices for AI Notable professions impacted are office workers, particularly those that work in data management, with estimates suggesting a nearly ** percent automation potential in data processing with generative AI. This means the field has begun to level, as lower educated professions and more manual labor-oriented professions are of far less risk to automation due to generative AI specifically.

Generative Artificial Intelligence (Gen AI) Market Outlook

The global Generative AI (Gen AI) market is valued at USD 38.06 billion in 2024 and is expanding at a compound annual growth rate (CAGR) of around 35%, reaching an estimated value of $200 billion by 2032.



Key segments contributing to this growth include software, which accounts for approximately 60% of the market share, and the healthcare and finance applications, which are forecasted to see the highest adoption rates. The cloud deployment mode will dominate with over 70% of the market share, reflecting the ongoing trend towards cloud-based solutions. Large enterprises will continue to lead in terms of enterprise size, while the Asia Pacific region is anticipated to exhibit the fastest growth, fuelled by rapid technological advancements and increasing investments in AI infrastructure.

The Generative AI market is set to experience significant growth driven by the continuous advancements in machine learning and deep learning technologies. As these AI models become more capable and efficient, they are being integrated into a broader array of business processes, enhancing productivity and innovation. The growing digital transformation across industries also propels the demand for AI capabilities, particularly in areas like customer experience management, predictive maintenance, and supply chain optimization. Additionally, the reduction in costs associated with AI technologies, due to improvements in cloud computing infrastructures and the democratization of AI tools, makes these technologies accessible to a wider range of businesses, including small and medium-sized enterprises. The global push towards more data-driven decision-making further amplifies the adoption and investment in Generative AI, underpinning its market growth.

Scope of the Generative Artificial Intelligence (Gen AI) Market Report

The market report includes an assessment of the market trends, segments, and regional markets. Overview and dynamics are included in the report.

Generative Ai Media Software is playing a pivotal role in transforming the media landscape by enabling the creation of highly realistic and engaging content. This software leverages advanced algorithms to generate images, videos, and even music, offering new possibilities for content creators and media companies. By automating parts of the creative process, Generative Ai Media Software allows for more efficient production workflows and the ability to personalize content at scale. This has led to a surge in innovative applications, such as virtual influencers and AI-generated characters, which are reshaping how audiences intera

Generative AI experienced a massive expansion of use cases in financial services during 2024, with customer experience and engagement emerging as the dominant application. A 2024 survey revealed that ** percent of respondents prioritized this area, a dramatic increase from ** percent in the previous year. Report generation, investment research, and document processing also gained significant traction, with over ** percent of firms implementing these applications. Additional use cases included synthetic data generation, code assistance, software development, marketing and sales asset creation, and enterprise research.

Snapshot img

AI Data Center Market Size 2025-2029

The AI data center market size is forecast to increase by USD 35.54 billion at a CAGR of 28.7% between 2024 and 2029.

The market is experiencing significant growth, driven by the explosion of generative AI and large language models. These advanced technologies demand immense computational power, leading to an increased focus on data centers as the backbone of AI infrastructure. A key trend in this market is the ubiquity of liquid cooling as a baseline requirement for high-performance data centers. This cooling technology enables more efficient heat dissipation and higher power densities, making it essential for data centers to meet the escalating demands of AI workloads. However, the market faces substantial challenges. IT service management and network security protocols are essential for maintaining system resilience and reliability.
As the energy requirements for AI processing continue to escalate, securing a reliable and sustainable power supply becomes a critical concern for market participants. Companies must navigate these challenges by exploring renewable energy sources, implementing energy storage solutions, and optimizing energy usage through advanced cooling technologies and power management systems. Virtual desktop infrastructure and remote access solutions enable secure and efficient access to applications and data from anywhere. By addressing these challenges and capitalizing on the opportunities presented by the growing demand for AI infrastructure, market players can effectively position themselves in the dynamic and evolving market.

What will be the Size of the AI Data Center Market during the forecast period?

Explore in-depth regional segment analysis with market size data - historical 2019-2023 and forecasts 2025-2029 - in the full report.
Request Free Sample

In the dynamic market, energy consumption reduction is a top priority, driving the adoption of data center design innovations such as precision cooling systems, liquid cooling technology, and airflow management. Performance benchmarks are crucial for selecting optimal AI infrastructure costs, while uninterruptible power supply and power monitoring tools ensure uptime and compliance with regulations. Power distribution units and capacity management systems enable the efficient use of renewable energy sources. Risk assessment methods and access control systems secure data, while data encryption techniques protect against cyber threats.

Compliance regulations, such as those related to environmental monitoring and waste heat recovery, are shaping the industry. Uptime monitoring, server consolidation, virtual desktop infrastructure, and rack-level monitoring optimize performance, and AI-driven analytics facilitate data center migration. Building management systems integrate various functions, including power distribution, environmental monitoring, and performance optimization, enhancing overall efficiency. Power scarcity and electrical grid constraints pose significant obstacles to the expansion of data centers.

How is this AI Data Center Industry segmented?

The AI data center industry research report provides comprehensive data (region-wise segment analysis), with forecasts and estimates in 'USD million' for the period 2025-2029, as well as historical data from 2019-2023 for the following segments.

Component

  Hardware
  Software
  Services


Type

  Hyperscale data centers
  Edge data centers
  Colocation Data centers


Deployment

  Cloud-based
  On-premises
  Hybrid cloud


Geography

  North America

    US
    Canada


  Europe

    France
    Germany
    The Netherlands
    UK


  APAC

    Australia
    China
    India
    Japan


  Rest of World (ROW)

By Component Insights

The Hardware segment is estimated to witness significant growth during the forecast period. The market is witnessing significant transformation, with the hardware segment leading the way. This segment includes the complete physical infrastructure designed for the high computational density required by artificial intelligence workloads. At its core are accelerators, specialized processors that handle the parallel mathematical operations necessary for training and inference. The market is heavily influenced by the product cycles of these components. For instance, the launch of NVIDIA's Blackwell architecture in March 2024 set a new performance benchmark, necessitating data center upgrades to accommodate its substantial power and cooling demands. Network security protocols are a critical concern as AI workloads increase, necessitating advanced cybersecurity measures.

Capacity forecasting is essential to ensure IT infrastructure management meets the demands of AI-powered applications. Cloud computing infrastructure is a significant trend, with many organizations opting for the flexibility and scalability it offers.

Comparison of Seconds to Output 500 Tokens, including reasoning model 'thinking' time; Lower is better by Model

Abstract:This dataset presents survey responses from first-year engineering students on their use of ChatGPT and other AI tools in a project-based learning environment. Collected as part of a study on AI’s role in engineering education, the data captures key insights into how students utilize ChatGPT for coding assistance, conceptual understanding, and collaborative work. The dataset includes responses on frequency of AI usage, perceived benefits and challenges, ethical concerns, and the impact of AI on learning outcomes and problem-solving skills.With AI increasingly integrated into education, this dataset provides valuable empirical evidence for researchers, educators, and policymakers interested in AI-assisted learning, STEM education, and academic integrity. It enables further analysis of student perceptions, responsible AI use, and the evolving role of generative AI in higher education.By making this dataset publicly available, we aim to support future research on AI literacy, pedagogy, and best practices for integrating AI into engineering and science curricula..................................................................................................................................................................Related PublicationThis dataset supports the findings presented in the following peer-reviewed article:ChatGPT in Engineering Education: A Breakthrough or a Challenge?Davood KhodadadPublished: 7 May 2025 | Physics Education, Volume 60, Number 4© 2025 The Author(s). Published by IOP Publishing LtdCitation: Davood Khodadad 2025 Phys. Educ. 60 045006DOI: 10.1088/1361-6552/add073If you use or reference this dataset, please consider citing the above publication......................................................................................................................................................................Description of the data and file structureTitle: ChatGPT in Engineering Education: Survey Data on AI Usage, Learning Impact, and CollaborationDescription of Data Collection:This dataset was collected through a survey distributed via the Canvas learning platform following the completion of group projects in an introductory engineering course. The survey aimed to investigate how students engaged with ChatGPT and other AI tools in a project-based learning environment, particularly in relation to coding, report writing, idea generation, and collaboration.The survey consisted of 15 questions:12 multiple-choice questions to capture quantitative insights on AI usage patterns, frequency, and perceived benefits.3 open-ended questions to collect qualitative perspectives on challenges, ethical concerns, and students' reflections on AI-assisted learning.Key areas assessed in the survey include:Students’ prior familiarity with AI tools before the course.Frequency and purpose of ChatGPT usage (e.g., coding assistance, conceptual learning, collaboration).Perceived benefits and limitations of using AI tools in an engineering learning environment.Ethical considerations, including concerns about over-reliance and academic integrity.The dataset provides valuable empirical insights into the evolving role of AI in STEM education and can support further research on AI-assisted learning, responsible AI usage, and best practices for integrating AI tools in engineering education.

Recent developments in generative large language models (LLMs) have raised questions about how this technology will affect higher education. We present results from two original surveys, collected in collaboration with the American Political Science Association, examining how instructors perceive the impact of generative LLMs. We find that educators are slightly pessimistic about generative LLMs, but support for AI tools varies based on application. Despite a professed importance for students to learn how to use AI tools, results show that educators' responses would primarily come through the prevention and detection of AI use rather than integrating AI into the curriculum. However, there are notable issues with detection and AI-ban enforcement. Respondents correctly determined whether students or AI wrote an essay in our essay bank no better than a coin flip; detection software are inaccurate. Based on these findings and suggestions from surveyed colleagues, we conclude with recommendations for dealing with generative AI in the classroom.

IntroductionProviding one-on-one support to large cohorts is challenging, yet emerging AI technologies show promise in bridging the gap between the support students want and what educators can provide. They offer students a way to engage with their course material in a way that feels fluent and instinctive. Whilst educators may have views on the appropriates for AI, the tools themselves, as well as the novel ways in which they can be used, are continually changing.MethodsThe aim of this study was to probe students' familiarity with AI tools, their views on its current uses, their understanding of universities' AI policies, and finally their impressions of its importance, both to their degree and their future careers. We surveyed 453 psychology and sport science students across two institutions in the UK, predominantly those in the first and second year of undergraduate study, and conducted a series of five focus groups to explore the emerging themes of the survey in more detail.ResultsOur results showed a wide range of responses in terms of students' familiarity with the tools and what they believe AI tools could and should not be used for. Most students emphasized the importance of understanding how AI tools function and their potential applications in both their academic studies and future careers. The results indicated a strong desire among students to learn more about AI technologies. Furthermore, there was a significant interest in receiving dedicated support for integrating these tools into their coursework, driven by the belief that such skills will be sought after by future employers. However, most students were not familiar with their university's published AI policies.DiscussionThis research on pedagogical methods supports a broader long-term ambition to better understand and improve our teaching, learning, and student engagement through the adoption of AI and the effective use of technology and suggests a need for a more comprehensive approach to communicating these important guidelines on an on-going basis, especially as the tools and guidelines evolve.

Generative AI Security Market Outlook

According to our latest research, the global Generative AI Security market size stood at USD 1.98 billion in 2024, reflecting robust momentum driven by the rapid integration of generative AI technologies across industries. The market is projected to expand at a CAGR of 28.1% from 2025 to 2033, reaching a forecasted value of USD 17.54 billion by 2033. This exceptional growth is underpinned by the escalating adoption of generative AI tools and the surging need for advanced security solutions to mitigate emerging AI-driven threats. As organizations increasingly leverage generative AI for innovation and automation, the imperative to secure these systems propels the market forward, making generative AI security a critical investment area for enterprises worldwide.

The primary growth driver for the generative AI security market is the exponential increase in the deployment of generative AI models across business processes and digital ecosystems. Organizations are leveraging generative AI for content creation, data analysis, and automation, but these advancements also introduce new vectors for cyber threats, such as data poisoning, model inversion, and adversarial attacks. The sophistication of these threats necessitates equally advanced security frameworks, prompting firms to invest in specialized generative AI security solutions. Moreover, the rising number of high-profile breaches involving AI-generated content and deepfakes has heightened awareness among both enterprises and regulators, further accelerating demand for robust generative AI security platforms.

Another significant factor fueling market growth is the tightening regulatory landscape surrounding AI and data security. Governments and industry bodies across North America, Europe, and Asia Pacific are introducing stringent compliance requirements to safeguard sensitive data processed by AI systems. These regulations mandate organizations to implement advanced security protocols, including real-time monitoring, threat detection, and automated response mechanisms specifically tailored for generative AI environments. Additionally, the growing emphasis on ethical AI usage and transparency compels organizations to adopt security solutions that not only protect data but also ensure the integrity and accountability of AI-generated outputs. This regulatory pressure, combined with increasing consumer expectations for privacy and trust, is a key catalyst for sustained market expansion.

The proliferation of cloud-based generative AI solutions is also reshaping the security landscape, creating both opportunities and challenges for market stakeholders. Cloud deployments offer scalability and flexibility, enabling organizations to rapidly experiment with and deploy generative AI models. However, this shift also exposes enterprises to new security risks, including multi-tenant vulnerabilities, data leakage, and unauthorized access to AI models and training data. As a result, there is a surge in demand for cloud-native generative AI security solutions that can provide end-to-end protection across distributed environments. Vendors are responding with innovations in secure model deployment, encryption, and access control, driving the evolution of the market and reinforcing the need for specialized expertise in generative AI security.

Regionally, North America continues to dominate the generative AI security market, accounting for the largest revenue share in 2024, followed closely by Europe and Asia Pacific. The United States leads in both adoption and innovation, supported by a mature technology ecosystem and proactive regulatory initiatives. Europe is witnessing rapid growth due to the enforcement of GDPR and AI Act regulations, while Asia Pacific is emerging as a high-growth region driven by digital transformation initiatives in China, Japan, and India. Each region presents unique opportunities and challenges, with local market dynamics, regulatory frameworks, and industry verticals shaping the trajectory of generative AI security adoption.

Component Analysis

The generative AI security market is segmented by component into software, hardware, and services, each playing a pivotal role in the overall security architecture. The software segment dominates the market, accounting for the highest revenue share in 2024, as organizations prioritize investment in advanced security platforms, threat detection tools, and AI-driven analytics. These software so

The Generative AI Smartphone market is poised for explosive growth, driven by advancements in AI processing capabilities, improved energy efficiency, and increasing consumer demand for enhanced mobile experiences. While precise market sizing data is unavailable, a reasonable estimate, considering the rapid adoption of AI in various sectors and the burgeoning smartphone market, suggests a 2025 market value of approximately $5 billion. This figure reflects the early-stage nature of generative AI integration into smartphones, with limited widespread adoption but significant potential for rapid expansion. We project a Compound Annual Growth Rate (CAGR) of 40% from 2025 to 2033, driven by factors such as the integration of larger language models directly into devices, enabling real-time translation, advanced image generation, personalized content creation, and enhanced user assistance. Key players like Samsung, Apple, Google, Oppo, OnePlus, Huawei, Vivo, Xiaomi, and HONOR are actively investing in research and development, fueling innovation and competition within this rapidly evolving market segment. The major restraining factors are currently high development costs, potential privacy concerns surrounding data usage, and the need for robust and energy-efficient hardware to support demanding AI algorithms. Market segmentation will likely evolve around processing power, AI features, price points, and target demographics. The forecast period (2025-2033) promises to be a defining era for the Generative AI Smartphone market. As technology matures, we anticipate a decline in production costs and a subsequent increase in affordability, accelerating market penetration. Further, continuous improvements in AI model efficiency will lead to longer battery life and better performance, addressing current limitations. The successful integration of advanced security protocols and robust data privacy measures will be crucial to address consumer concerns and promote market growth. Ultimately, the long-term success of this market hinges on the ability of manufacturers to deliver innovative applications that seamlessly integrate generative AI into the everyday user experience, providing tangible benefits and value beyond mere novelty.

Generative AI Application Market Outlook

The global Generative AI application market size was estimated at USD 10.7 billion in 2023 and is projected to reach USD 205.9 billion by 2032, expanding at a compound annual growth rate (CAGR) of 39.7% during the forecast period. The proliferation of advanced AI technologies and the integration of machine learning (ML) and deep learning (DL) models are primarily driving the growth of this market.

One of the primary growth factors contributing to the generative AI market is the escalating demand for automation across various industries. Businesses are increasingly adopting generative AI to optimize operational efficiencies, reduce costs, and enhance customer experiences. The healthcare sector, for example, is leveraging AI to improve diagnostic accuracy, personalize treatment plans, and streamline administrative processes. Such innovations are expected to enhance patient outcomes and significantly reduce healthcare costs.

Another critical element fueling market growth is the rapid advancements in computing power and data storage capabilities. With the advent of high-performance computing (HPC) systems and cloud-based platforms, organizations can now process and analyze large volumes of data at unprecedented speeds. This capability is crucial for the effective deployment of generative AI models that require immense computational resources. Moreover, the decreasing costs of hardware and cloud services are making these technologies more accessible to small and medium enterprises (SMEs), thereby broadening the market scope.

The increasing investments in AI research and development by both public and private sectors are also playing a pivotal role in market expansion. Governments around the globe are launching initiatives and funding programs to bolster AI capabilities, aiming to secure their positions as leaders in the global AI race. Concurrently, private companies are pouring substantial capital into AI startups and research projects, accelerating innovation and commercialization of generative AI applications. These concerted efforts are anticipated to drive significant advancements in AI technologies, further propelling market growth.

Artificial Intelligence (AI) Verticals are becoming increasingly significant as industries seek to harness the power of AI to address specific challenges and opportunities. These verticals refer to specialized sectors where AI technologies are applied to solve unique problems or enhance processes. For instance, in the healthcare vertical, AI is used for predictive analytics and personalized medicine, while in finance, it aids in algorithmic trading and risk management. The development of AI verticals allows for tailored solutions that meet the distinct needs of different industries, thereby driving innovation and efficiency. As AI continues to evolve, the expansion of AI verticals is expected to create new opportunities for businesses to leverage AI in a more targeted and effective manner.

Regionally, North America is poised to dominate the generative AI market owing to the presence of numerous tech giants and a robust ecosystem for AI research and development. However, Asia Pacific is expected to witness the highest growth rate during the forecast period, driven by increasing investments in AI technologies, burgeoning startup ecosystems, and supportive government policies. Europe and Latin America are also emerging as potential markets due to their growing focus on digital transformation and AI adoption across various sectors.

Component Analysis

The generative AI application market can be broadly segmented based on components into software, hardware, and services. The software segment holds the largest market share, owing to the extensive usage of AI algorithms, natural language processing (NLP) tools, and machine learning frameworks. AI software applications are being widely adopted across industries to automate processes, gain insights from data, and enhance decision-making capabilities. The continuous advancements in AI software tools and platforms are expected to drive this segment's growth significantly over the forecast period.

The hardware segment, encompassing GPUs, TPUs, and other specialized AI processors, is also witnessing substantial growth. The increasing complexity and computational demands of AI models necessitate the use of high-performance hardware. Companies are investing heavily in AI-sp

This replication package provides the setup and results to generate OpenJML code contracts for Java source code by fine-tuning and employing the resulting CodeT5 and CodeT5+ transformer models. Our code contract generation setup involved the training of the AI models and application. Furthermore, we analyzed the generated annotations wrt. thier logical validity and the type of OpenJML compilation errors. Both methods, together with the results are similarly provided.

Source Code Repository (see also scripts-sources.tar):

• https://github.com/SEG-UNIBE/auto-generated-code-contracts
• https://zenodo.org/doi/10.5281/zenodo.13356451

Replication Package: contains the following [folders]

• Scripts:
  • [scripts-sources.tar]: source codes of the following scripts
    • Python scripts that we used for training and adding the OpenJML code contracts to the Java methods
    • automated analyses of the studied source code classes and the type of compilation errors
• Sourcegraph Search Results:
  • [sourcegraph-results.tar]: the results of the Sourcegraph search queries
• Datasets:
  • [dataset.tar]: the dataset including the weka-project which contributes two-thirds of the contracts
  • [dataset-withoutweka.tar]: the dataset without weka, which is significantly smaller and was used to examine the performance bias when training and testing without weka
• CodeT5 Models:
  
  • [codet5-contracts.tar]: the best performing CodeT5 model which was fine-tuned to create OpenJML annotations for methods
  • [codet5p-contracts.tar]: the best performing CodeT5+ model which was fine-tuned to create OpenJML annotations for methods
  • [codet5p-contracts-withoutweka.tar]: the CodeT5+ model which was trained without weka on the same task
• Analysis Results:
  • [analysis-results.tar/compilability-analysis]: the results of the compilability analysis
    • the subjects to which we applied the best performing CodeT5+
    • the compilation results and their analysis
  • [analysis-results.tar/logical-analysis] the results of the logical analysis
    • the analysis of logic validity of SimpleStack and SimpleTicTacToe

Generative AI Technology Market Outlook

The global generative AI technology market size is projected to grow exponentially, with a market value reaching approximately $40 billion by 2032 from $8 billion in 2023, reflecting a robust compound annual growth rate (CAGR) of 18.5%. The primary factor driving this growth is the increasing demand for AI-driven solutions that enhance productivity and innovation across various sectors.

The surge in demand for generative AI solutions is predominantly being driven by advancements in machine learning algorithms and the increasing volume of data generated globally. Organizations are recognizing the potential of AI to automate complex tasks, generate creative content, and provide deep insights from vast datasets, thereby leading to a significant reduction in operational costs and improvement in efficiency. The implementation of AI technologies is transforming industries, enabling new applications in fields such as drug discovery, automated content creation, and personalized marketing.

Another critical growth factor is the integration of AI with other emerging technologies such as the Internet of Things (IoT), blockchain, and cloud computing. The convergence of these technologies is creating new opportunities for innovation and enhanced capabilities in data analysis, cybersecurity, and smart automation. For instance, AI-powered IoT devices are becoming increasingly popular in sectors such as healthcare and manufacturing, where they contribute to predictive maintenance, remote monitoring, and enhanced decision-making processes.

Furthermore, the proliferation of AI research and development initiatives, supported by substantial investments from both private enterprises and government bodies, is accelerating the growth of the generative AI market. Countries across the globe are developing strategic plans to foster AI innovation, aiming to become leaders in the AI ecosystem. These initiatives are not only providing financial support but also creating a conducive environment for startups and established companies to explore and expand AI capabilities.

Ai Face Generators are a fascinating development within the realm of generative AI technologies, offering new possibilities for creative expression and practical applications. These generators use advanced algorithms to create realistic human faces, which can be utilized in various industries such as entertainment, gaming, and marketing. By synthesizing human-like features, AI face generators can produce avatars and virtual characters that enhance user engagement and provide personalized experiences. Moreover, they are being explored for use in identity verification systems, where they can improve security measures by generating unique facial features. As the technology continues to evolve, ethical considerations around privacy and consent are becoming increasingly important, prompting discussions on how to responsibly integrate AI face generators into society.

Geographically, North America holds the largest share of the generative AI market, attributed to the presence of leading technology companies and research institutions. The region's advanced infrastructure and high adoption rate of AI technologies across various industries further bolster its market position. Asia Pacific is expected to witness the highest growth rate during the forecast period, driven by rapid digital transformation, increasing investments in AI, and supportive government policies. Europe, Latin America, and the Middle East & Africa are also anticipated to experience considerable growth, although at varying paces depending on the region's technological maturity and economic conditions.

Software Analysis

The software segment constitutes a significant portion of the generative AI market, encompassing various tools and platforms that facilitate the creation and implementation of AI models. This segment includes applications such as natural language processing (NLP), computer vision, and generative adversarial networks (GANs), which are instrumental in developing AI-driven solutions. The increasing adoption of AI software in sectors like healthcare, finance, and media is driving the demand for sophisticated AI tools that can generate high-quality content and provide valuable insights.

One of the critical drivers for the software segment is the growing need for automation in business processes. Many organizations are leveraging AI software to autom

Data from "A comparison of Human, GPT-3.5, and GPT-4 Performance in a University-Level Coding Course". This study evaluates the performance of ChatGPT variants, GPT-3.5 and GPT-4, both with and without prompt engineering, against solely student work and a mixed category containing both student and GPT-4 contributions in university-level physics coding assignments using the Python language. Comparing 50 student submissions to 50 AI-generated submissions across different categories, and marked blindly by three independent markers, we amassed n = 300 data points. Students averaged 91.9% (SE:0.4), surpassing the highest performing AI submission category, GPT-4 with prompt engineering, which scored 81.1% (SE:0.8) - a statistically significant difference (p = 2.482 x 10-10). Prompt engineering significantly improved scores for both GPT-4 (p = 1.661 x 10-4) and GPT-3.5 (p = 4.967 x 10-9). Additionally, the blinded markers were tasked with guessing the authorship of the submissions on a four-point Likert scale from Definitely AI' toDefinitely Human'. They accurately identified the authorship, with 92.1% of the work categorized as 'Definitely Human' being human-authored. Simplifying this to a binary AI' orHuman' categorization resulted in an average accuracy rate of 85.3%. These findings suggest that while AI-generated work closely approaches the quality of university students' work, it often remains detectable by human evaluators.

Claude 2, developed by rising startup star Anthropic, is the most capable large language model generative AI on the current market. It reached a success ratio of ** percent with the HumanEval benchmark. This is particularly noteworthy as it is a 0-shot evaluation, meaning all AI programs benchmarked against it had not had previous data of this sort nor previous training with the tasks. This means that Claude 2 was the quickest at absorbing and understanding the task given to it.

Geoparsing with Large Language Models

The .zip file included in this repository contains all the code and data required to reproduce the results from our paper. Note, however, that in order to run the OpenAI models, users will required an OpenAI API key and sufficient API credits.

Data

The data used for the paper are in the datasetst and results folders.

**Datasets: **This contains the XML files (LGL and Geovirus) and Json files (News2024) used to benchmark the models. It also contains all the data used to fine-tune the gpt-3.5 model, the prompt templates sent to the LLMs, and other data used for mapping and data creation.

**Results: **This contains the results for the models on the three datastes. The folder is separated by dataset, with a single .csv file giving the results for each model on each dataset separately. The .csv file is structured so that each row contains either a predicted toponym and an associated true toponym (along with assigned spatial coordinates), if the model correctly identified a toponym; otherwise the true toponym columns are empty for false positives and the predicted columns are empty for false negatives.

Code

The code is split into two seperate folders gpt_geoparser and notebooks.

**GPT_Geoparser: **this contains the classes and methods used process the XML and JSON articles (data.py), interact with the Nominatim API for geocoding (gazetteer.py), interact with the OpenAI API (gpt_handler.py), process the outputs from the GPT models (geoparser.py) and analyse the results (analysis.py).

Notebooks: This series of notebooks can be used to reproduce the results given in the paper. The file names a reasonably descriptive of what they do within the context of the paper.

Code/software

Requirements

Numpy

Pandas

Geopy

Scitkit-learn

lxml

openai

matplotlib

Contextily

Shapely

Geopandas

tqdm

huggingface_hub

Gnews

Access information

Other publicly accessible locations of the data:

The LGL and GeoVirus datasets can also be obtained here (opens in new window).

Abstract

Geoparsing- the process of associating textual data with geographic locations - is a key challenge in natural language processing. The often ambiguous and complex nature of geospatial language make geoparsing a difficult task, requiring sophisticated language modelling techniques. Recent developments in Large Language Models (LLMs) have demonstrated their impressive capability in natural language modelling, suggesting suitability to a wide range of complex linguistic tasks. In this paper, we evaluate the performance of four LLMs - GPT-3.5, GPT-4o, Llama-3.1-8b and Gemma-2-9b - in geographic information extraction by testing them on three geoparsing benchmark datasets: GeoVirus, LGL, and a novel dataset, News2024, composed of geotagged news articles published outside the models' training window. We demonstrate that, through techniques such as fine-tuning and retrieval-augmented generation, LLMs significantly outperform existing geoparsing models. The best performing models achieve a toponym extraction F1 score of 0.985 and toponym resolution accuracy within 161 km of 0.921. Additionally, we show that the spatial information encoded within the embedding space of these models may explain their strong performance in geographic information extraction. Finally, we discuss the spatial biases inherent in the models' predictions and emphasize the need for caution when applying these techniques in certain contexts.

Methods

This contains the data and codes required to reproduce the results from our paper. The LGL and GeoVirus datasets are pre-existing datasets, with references given in the manuscript. The News2024 dataset was constructed specifically for the paper.

To construct the News2024 dataset, we first created a list of 50 cities from around the world which have population greater than 1000000. We then used the GNews python package https://pypi.org/project/gnews/ (opens in new window) to find a news article for each location, published between 2024-05-01 and 2024-06-30 (inclusive). Of these articles, 47 were found to contain toponyms, with the three rejected articles referring to businesses which share a name with a city, and which did not otherwise mention any place names.

We used a semi autonmous approach to geotagging the articles. The articles were first processed using a Distil-BERT model, fine tuned for named entity recognicion. This provided a first estimate of the toponyms within the text. A human reviewer then read the articles, and accepted or rejected the machine tags, and added any tags missing from the machine tagging process. We then used OpenStreetMap to obtain geographic coordinates for the location, and to identify the toponym type (e.g. city, town, village, river etc). We also flagged if the toponym was acting as a geo-political entity, as these were reomved from the analysis process. In total, 534 toponyms were identified in the 47 news articles.

A presentation at Australasian STEM / Engineering Librarians Virtual Workshop on 13 June 2024As generative AI tools become more prevalent in academic research, STEM libraries play vital role in promoting their responsible and ethical use. This discussion delves into how libraries can facilitate the adoption of AI technologies such as literature discovery tools, data analysis platforms, and writing assistants, while addressing key challenges that academic libraries face. Effective utilisation of these tools requires collaboration and coordination among librarians, researchers, IT professionals, and other stakeholders.The potential benefits of generative AI in boosting productivity and research capabilities are examined. At the same time, risks such as accuracy concerns, authorship ambiguity, threats to academic integrity (including plagiarism), and over-reliance on AI are critically assessed.Strategies for STEM libraries are discussed, including offering training on responsible AI use, establishing citation guidelines, and reinforcing academic integrity policies. Case studies and examples of best practices illustrate how to leverage generative AI while maintaining rigorous academic standards.By fostering a nuanced understanding of the opportunities and challenges presented by generative AI, libraries can equip researchers with the knowledge and support needed to use these tools effectively. This presentation aims to empower STEM librarians and researchers to proactively shape the responsible integration of generative AI into the research process, ensuring it augments, rather than undermines, human expertise and academic rigour.

This dataset supports a literature mapping of AI-driven content generation, analyzing 631 solutions published over the last five years to better understand and characterize the Generative Artificial Intelligence landscape. Tools like ChatGPT, Dall-E, or Midjourney have democratized access to Large Language Models, enabling the creation of human-like content. However, the concept 'Generative Artificial Intelligence' lacks a universally accepted definition, leading to potential misunderstandings.

The study has been published in International Journal of Interactive Multimedia and Artificial Intelligence.

García-Peñalvo, F. J., & Vázquez-Ingelmo, A. (2023). What do we mean by GenAI? A systematic mapping of the evolution, trends, and techniques involved in Generative AI. International Journal of Interactive Multimedia and Artificial Intelligence, In Press.