Automated documentation of programming source code is a challenging task with significant practical and scientific implications for the developer community. We present a large language model (LLM)-based application that developers can use as a support tool to generate basic documentation for any publicly available repository. Over the last decade, several papers have been written on generating documentation for source code using neural network architectures. With the recent advancements in LLM technology, some open-source applications have been developed to address this problem. However, these applications typically rely on the OpenAI APIs, which incur substantial financial costs, particularly for large repositories. Moreover, none of these open-source applications offer a fine-tuned model or features to enable users to fine-tune. Additionally, finding suitable data for fine-tuning is often challenging. Our application addresses these issues which is available at https://pypi.org/project/readme-ready/.

Upon release, ChatGPT3.5 shocked the software engineering community by its ability to generate answers to specialized questions about coding. Immediately, many educators wondered if it was possible to use the chatbot as a support tool that helps students answer their programming questions. This article evaluates this possibility at three levels: fundamental Computer Science knowledge (basic algorithms and data structures), core competency (design patterns), and advanced knowledge (quantum computing). In each case, we ask normalized questions several times to ChatGPT3.5, then look at the correctness of answers, and finally check if this creates issues. The main result is that the performances of ChatGPT3.5 degrades drastically as the specialization of the domain increases: for basic algorithms it returns answers that are almost always correct, for design patterns the generated code contains many code smells and is generally of low quality, but it is still sometimes able to fix it (if asked), and for quantum computing it is often blatantly wrong.

The integration of artificial intelligence (AI) and optimization is transforming the landscape of engineered systems, offering unprecedented opportunities to enhance efficiency, reliability, and resilience across domains (Palle, 2023) such as power systems (Thirunavukkarasu et al., 2023), supply chains, and logistics (Joel et al., 2024). As these networked systems become more dependent on AI-enabled decision support tools, the ethical challenges associated with their deployment grow more complex (Whittlestone and Clarke, 2022). Traditional ethical concerns in AI--such as fairness, accountability, and transparency--take on new dimensions when applied to systems characterized by complex networks and optimization processes, where decisions have far-reaching societal impacts (Jobin et al., 2019). Governments and organizations worldwide have responded to these ethical concerns by introducing frameworks and regulations aimed at ensuring trustworthy AI (Harrison and Luna-Reyes, 2022; Weaver, 2021; Aoki et al., 2024; Madhavan et al., 2020). Initiatives like the European Union's AI Act (Parliament and of the European Union, 2024) and the Biden-Harris administration's AI Bill of Rights (Biden, 2021) aim to safeguard fairness, transparency, and accountability in AI systems (White House Office of Science and Technology Policy, 2023; OECD, 2020; Radu, 2021).

The natural language capabilities demonstrated by large language models (LLMs) highlight an opportunity for new writing support tools that address the varied needs of people with dyslexia. We present LaMPost, a prototype email editor that draws upon our understanding of these needs to motivate AI-powered writing features, such as outlining main ideas, generating a subject line, suggesting changes, and rewriting a selection. We evaluated LaMPost with 19 adults with dyslexia, identifying promising routes for further exploration (such as the popular "rewrite" and "subject line" features), while also finding that the current generation of LLMs may not yet meet the accuracy and quality thresholds to be useful for writers with dyslexia. In addition, knowledge of the AI did not alter participants' perception of the system nor their feelings of autonomy, expression, and self-efficacy when writing emails. Our findings provide insight into the benefits and drawbacks of LLMs as writing support for adults with dyslexia, and they offer a foundation to build upon in future research. Dyslexia refers to a cluster of symptoms that result in challenges with word recognition, reading fluency, spelling, and writing population.8

We diagnose and briefly discuss the dearth of the author: a condition that arises when AI-based creativity support tools for writing allow users to produce large amounts of text without making a commensurate number of creative decisions, resulting in output that is sparse in expressive intent. We argue that the dearth of the author helps to explain a number of recurring difficulties and anxieties around AI-based writing support tools, but that it also suggests an ambitious new goal for AI-based CSTs.

Dyslexia is the most widespread specific learning disorder and significantly impair different cognitive domains. This, in turn, negatively affects dyslexic students during their learning path. Therefore, specific support must be given to these students. In addition, such a support must be highly personalized, since the problems generated by the disorder can be very different from one to another. In this work, we explored the possibility of using AI to suggest the most suitable supporting tools for dyslexic students, so as to provide a targeted help that can be of real utility. To do this, we relied on recommendation algorithms, which are a branch of machine learning, that aim to detect personal preferences and provide the most suitable suggestions. We hence implemented and trained three collaborative-filtering recommendation models, namely an item-based, a user-based and a weighted-hybrid model, and studied their performance on a large database of 1237 students' information, collected with a self-evaluating questionnaire regarding all the most used supporting strategies and digital tools. Each recommendation model was tested with three different similarity metrics, namely Pearson correlation, Euclidean distance and Cosine similarity. The obtained results showed that a recommendation system is highly effective in suggesting the optimal help tools/strategies for everyone. This demonstrates that the proposed approach is successful and can be used as a new and effective methodology to support students with dyslexia.

Learning disorders are neurological conditions that affect the brain's ability to interconnect communication areas. Dyslexic students experience problems with reading, memorizing, and exposing concepts; however the magnitude of these can be mitigated through both therapies and the creation of compensatory mechanisms. Several efforts have been made to mitigate these issues, leading to the creation of digital resources for students with specific learning disorders attending primary and secondary education levels. Conversely, a standard approach is still missed in higher education. The VRAIlexia project has been created to tackle this issue by proposing two different tools: a mobile application integrating virtual reality (VR) to collect data quickly and easily, and an artificial intelligencebased software (AI) to analyze the collected data for customizing the supporting methodology for each student. The first one has been created and is being distributed among dyslexic students in Higher Education Institutions, for the conduction of specific psychological and psychometric tests. The second tool applies specific artificial intelligence algorithms to the data gathered via the application and other surveys. These AI techniques have allowed us to identify the most relevant difficulties faced by the students' cohort. Our different models have obtained around 90\% mean accuracy for predicting the support tools and learning strategies.

Institutions considering allowing students to opt out of data sharing should consider very carefully whether this may create or further amplify inequities faced by learners. Known as consent bias, the problem is that those students who choose to opt out (or decide not to opt in) may differ systematically, such that the conclusions or actions taken based on the data will unfairly bias one of the groups of students. At the moment, students generally don't feel they can control access to the data their college collects about them. According to the Student Voice survey conducted this summer by Inside Higher Ed and College Pulse, with support from Kaplan, only 22 percent of students believed they could restrict access to this, while 9 percent did not and the vast majority -- 69 percent -- weren't sure. Student Voice explores higher education from the perspective of students, providing unique insights on their attitudes and opinions. Kaplan provides funding and insights to support Inside Higher Ed's coverage of student polling data from College Pulse.

Speech recognition is definitely one of the areas where artificial intelligence is showing its power and effectiveness. And what is the last thing that journalists, secretaries, and assistants wish to take care of? But whether for interviews or parliamentary reports, new AI-based applications emerge as useful support tools to let the machine do the boring part of the job and allow people to focus on more demanding and intellectually challenging tasks. In the next year, the Estonian Parliament (Riigikogu) is set to introduce HANS – AI system that will be a valuable ally to the work of lawmakers and employees of the Riigikogu. By deploying speech recognition, it will increase the efficiency and accuracy in transcripts of the sessions.

An artificial intelligence (AI) system accurately identified key findings in chest X-rays of patients in the emergency department suspected of having pneumonia in just 10 seconds, researchers from Intermountain Healthcare and Stanford University reported at the European Respiratory Society's International Congress 2019. Traditionally, it takes physicians 20 minutes or more to identify pneumonia from chest X-rays. "In this initial study, we've demonstrated the algorithm's potential by validating it on patients in the emergency departments at Intermountain Healthcare," said Jeremy Irvin, a Ph.D. student at Stanford. "Our hope is that the algorithm can improve the quality of pneumonia care at Intermountain, from improving diagnostic accuracy to reducing time to diagnosis." Early diagnosis could lead to treatment starting earlier, which is vital for severely ill patients, the researchers noted.