Category
489 papers
Collaborative AI experimentation across industry and academia requires platforms that enable rapid prototyping while preserving controlled access, tenant separation, and transparent workflows. Despite growing interest in AI sandboxes, there is still limited practical guidance on how to design and implement platforms that integrate experimentation capabilities with governance requirements. This work presents the design and implementation of a governance-aware, multi-tenant AI sandbox for structured experimentation and the generation of reusable evaluation evidence across projects and stakeholder groups. The sandbox was developed within an industry-academia collaboration based on requirements that were iteratively refined with industrial partners. Its reference architecture separates the multi-tenant user interface from the backend control plane and places execution and data-management functions in dedicated layers. The platform supports governed user onboarding, project-centered collaboration, managed access to AI services, approval workflows, audit logging, and traceable experimentation. Experiment configurations, contextual information, and governance decisions are stored as persistent records, allowing evidence and outcomes to be compared and reused across projects. The development process provides practical lessons for deploying and extending governance-aware AI sandbox platforms in collaborative research and industrial environments.
Recent advances in agentic frameworks have enabled AI agents to perform complex reasoning and decision-making. However, evidence comparing their reasoning performance, efficiency, and practical suitability remains limited. To address this gap, we empirically evaluate 22 widely used agentic frameworks across three reasoning benchmarks: BBH, GSM8K, and ARC. The frameworks were selected from 1,200 GitHub repositories collected between January 2023 and July 2025 and organized into a taxonomy based on architectural design. We evaluated them under a unified setting, measuring reasoning accuracy, execution time, computational cost, and cross-benchmark consistency. Our results show that 19 of the 22 frameworks completed all three benchmarks. Among these, 12 showed stable performance, with mean accuracy of 74.6-75.9%, execution time of 4-6 seconds per task, and cost of 0.14-0.18 cents per task. Poorer results were mainly caused by orchestration problems rather than reasoning limits. For example, Camel failed to complete BBH after 11 days because of uncontrolled context growth, while Upsonic consumed USD 1,434 in one day because repeated extraction failures triggered costly retries. AutoGen and Mastra also exhausted API quotas through iterative interactions that increased prompt length without improving results. We also found a sharp drop in mathematical reasoning. Mean accuracy on GSM8K was 44.35%, compared with 89.80% on BBH and 89.56% on ARC. Overall, this study provides the first large-scale empirical comparison of agentic frameworks for reasoning-intensive software engineering tasks and shows that framework selection should prioritize orchestration quality, especially memory control, failure handling, and cost management.
Large Language Models (LLM) and Generative Pre-trained Transformers (GPT), are reshaping the field of Software Engineering (SE). They enable innovative methods for executing many software engineering tasks, including automated code generation, debugging, maintenance, etc. However, only a limited number of existing works have thoroughly explored the potential of GPT agents in SE. This vision paper inquires about the role of GPT-based agents in SE. Our vision is to leverage the capabilities of multiple GPT agents to contribute to SE tasks and to propose an initial road map for future work. We argue that multiple GPT agents can perform creative and demanding tasks far beyond coding and debugging. GPT agents can also do project planning, requirements engineering, and software design. These can be done through high-level descriptions given by the human developer. We have shown in our initial experimental analysis for simple software (e.g., Snake Game, Tic-Tac-Toe, Notepad) that multiple GPT agents can produce high-quality code and document it carefully. We argue that it shows a promise of unforeseen efficiency and will dramatically reduce lead-times. To this end, we intend to expand our efforts to understand how we can scale these autonomous capabilities further.
Reach audiences
Advertise in front of researchers, engineers, and readers.
The growing domain of liquidity in computing extends its boundaries to include advancements like liquid artificial intelligence (AI). Liquid AI leverages liquid software using isomorphic Internet of Things (IoT) architecture to enhance computation at the edge. This innovation unveils vast opportunities yet also introduces significant challenges, particularly around privacy and trust. We explore the vulnerabilities that might hinder the progression of this technological fusion toward achieving trustworthy AI. Through an intensive examination of the literature, this research highlights the heightened threats to data integrity and stakeholder trust in these evolving ecosystems. Four main challenges: Data collection, Data storage and Access, Data utilization and sharing, and Surveillance and profiling were identified and examined under privacy, and two, Algorithms and decision-making and Security of IoT infrastructure under trust. The concerns are further categorized to highlight their impact on the development of trustworthy AI. The study acknowledges the early state of the field. Consequently, this research navigates through the limited available literature, initiating a pioneering discourse emphasizing fostering a foundation for developing secure and trustworthy Liquid AI environments.
The rapid adoption of Generative AI (GenAI) in the software development life cycle (SDLC) increases computational demand, which can raise the carbon footprint of development activities. At the same time, organizations are increasingly embedding governance mechanisms into GenAI-assisted development to support trust, transparency, and accountability. However, these governance mechanisms introduce additional computational workloads, including repeated inference, regeneration cycles, and expanded validation pipelines, increasing energy use and the carbon footprint of GenAI-assisted development. This paper proposes Carbon-Aware Governance Gates (CAGG), an architectural extension that embeds carbon budgets, energy provenance, and sustainability-aware validation orchestration into human-AI governance layers. CAGG comprises three components: (i) an Energy and Carbon Provenance Ledger, (ii) a Carbon Budget Manager, and (iii) a Green Validation Orchestrator, operationalized through governance policies and reusable design patterns.
LLMs can imitate how people write, which raises concerns about impersonation, trust, and detection in social settings. These concerns are especially important for adolescents, who use generative AI frequently but may struggle to recognize it. We introduce \textit{DoppelBot}, a cooperative social deduction game designed to study how young people detect and respond to AI impersonation. Through studies with middle schoolers, we investigate whether a DoppelBot prompts reflection on privacy and impersonation, how repeated exposure affects AI-detection accuracy as agents become more personalized, and which strategies students use to identify AI doppelg\"angers. We find that students'detection accuracy improves over time, driven by a shift from relying on linguistic cues to leveraging shared social and contextual signals. Students also demonstrated an understanding of AI limitations such as embodiment and reflected on broader issues such as data privacy. To support future research, we release an anonymized dataset of game transcripts and voting behavior.
Artificial Intelligence-Generated Content (AIGC) services employ Generative AI (GenAI) models to automatically generate diverse content. Mobile AIGC networks host GenAI models on edge-located AIGC Service Providers (ASPs) to deliver low-latency and personalized AIGC services for mobile users. However, AIGC inference tasks typically occupy GPU memory until task completion, causing GPU memory exhaustion at serving ASPs and triggering out-of-memory failures rather than merely increasing service latency. Existing studies on AIGC task orchestration have largely overlooked GPU memory feasibility constraints. To address this issue, we develop a cooperative multi-agent orchestration framework, in which each edge node is equipped with a scheduling agent to route tasks to local ASPs or neighboring edge nodes. Since scheduling agents make decisions based only on local observations, while peer offloading couples their resource states and long-term utilities, we formulate the orchestration process as a cooperative Decentralized Partially Observable Markov Decision Process (Dec-POMDP). To solve the Dec-POMDP, we propose an \underline{A}ttention-guided \underline{M}ulti-\underline{A}gent deep reinforcement learning algorithm with \underline{Diff}usion policies (A-MADiff) under the centralized training with a decentralized execution paradigm. A-MADiff employs diffusion-based decentralized actors to generate multi-modal preferences over feasible orchestration actions, and an attention-guided centralized critic to estimate per-agent values from cross-agent states under GPU memory heterogeneity. Numerical results demonstrate that A-MADiff significantly improves the cumulative reward over the state-of-the-art baseline.
American Students are increasingly using generative AI to create inappropriate images and videos–known as deepfake non-consensual intimate images (NCII)–of young girls in the K-12 school system (Walker, 2025). To date, schools do not have clear guidelines on how to handle this rising crisis. Federal policy attention, such as the Take It Down Act, prioritizes removing the harmful content after it is put online, as opposed to taking steps to prevent deepfake NCII from being created. It is time to readjust how generative AI content is regulated. This brief examines different international approaches to mitigating the risk of AI violence against women, such as the AI Act and Digital Services Act (DSA), to propose preventative measures within U.S. policy. Preventing deepfake NCII from circulating schools would include the following: mandating that the NIST AI Risk Management Framework be adopted by all generative AI companies operating within the United States; defining AI deepfake NCII as a form of abuse; establishing rules stating what minors can access on the internet as well as creating safe, protected spaces for them to responsibly use the internet; implementing mandated free training for school teachers and administrations across the country about what constitutes proper AI usage and how to prevent AI-based violence; designing lessons for school-age children about the harm of deepfake NCII; expanding counseling services within public schools to assist children who are affected by deepfake NCII.
American Students are increasingly using generative AI to create inappropriate images and videos–known as deepfake non-consensual intimate images (NCII)–of young girls in the K-12 school system (Walker, 2025). To date, schools do not have clear guidelines on how to handle this rising crisis. Federal policy attention, such as the Take It Down Act, prioritizes removing the harmful content after it is put online, as opposed to taking steps to prevent deepfake NCII from being created. It is time to readjust how generative AI content is regulated. This brief examines different international approaches to mitigating the risk of AI violence against women, such as the AI Act and Digital Services Act (DSA), to propose preventative measures within U.S. policy. Preventing deepfake NCII from circulating schools would include the following: mandating that the NIST AI Risk Management Framework be adopted by all generative AI companies operating within the United States; defining AI deepfake NCII as a form of abuse; establishing rules stating what minors can access on the internet as well as creating safe, protected spaces for them to responsibly use the internet; implementing mandated free training for school teachers and administrations across the country about what constitutes proper AI usage and how to prevent AI-based violence; designing lessons for school-age children about the harm of deepfake NCII; expanding counseling services within public schools to assist children who are affected by deepfake NCII.
Evaluating synthetic data produced by generative models remains a critical challenge in sensitive domains such as healthcare and finance. Ensuring that such data is ‘faithful’ to real data is essential for downstream applications and decision-making, including regulatory compliance. This paper introduces an AI-powered interactive visual system—AISyst—designed to assess the fidelity of synthetic tabular datasets. The system supports multilevel comparisons with real datasets, spanning multivariate resemblance analyses based on dimensionality reduction through suitable two-dimensional projections, bivariate correlation and univariate similarity. AISyst also integrates an AI assistant by leveraging state-of-the-art large language models to summarize key findings and generate suggestions for improving synthetic data generation models. We validated the capabilities of AISyst through three case studies, supported by feedback from industrial AI experts who endorsed its broader deployment.
Anatomy education serves as the cornerstone of medical training across both modern and traditional medicine systems. However, traditional pedagogical approaches—relying heavily on cadaveric dissection and two-dimensional illustrations—face mounting challenges, including cadaver shortages, ethical concerns, and the need to engage technology-oriented Generation Z learners. Artificial Intelligence (AI) has emerged as a transformative force in anatomical education, offering virtual dissection simulations, adaptive learning platforms, intelligent tutoring systems, and generative AI-powered chatbots. In modern medical education, AI tools such as ChatGPT, Anatbuddy, and VR/AR-based platforms have demonstrated significant potential in personalizing learning, generating assessment materials, and enhancing student engagement. Concurrently, in Ayurvedic anatomy education (Sharira Rachana), AI-powered tools like CADAVIZ and AyurSIM are bridging traditional knowledge with contemporary technological methods, enabling three-dimensional visualization of anatomical concepts and addressing resource disparities. This review consolidates existing evidence on AI applications in both streams of anatomy education, evaluates their effectiveness, and identifies key challenges including content accuracy concerns, over-reliance on technology, ethical considerations, and the need for customized knowledge bases. The findings underscore that AI should augment rather than replace traditional teaching methods, with a balanced, ethically guided approach being essential for effective integration. Future directions include developing specialized AI tools for traditional medicine anatomy, integrating real clinical cases, and establishing systematic AI literacy programmes for both educators and students.
MIT affiliates engage with the MIT-IBM Computing Research Lab to bring rigorous theory to production systems.
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.
Earth AI
Gemini Omni 1.1 Flash brings a new suite of creative controls and generative video capabilities to developers.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
