Chatter about artificial intelligence mostly falls into three basic categories: anxious uncertainty (will it take our jobs?); In this hazy, liminal, pre-disruption moment, there is little consensus as to whether generative AI is a tool or a threat, and few rules for using it properly. For students, this uncertainty feels especially profound. Bans on AI and claims that using it constitutes cheating are now giving way to concerns that AI use is inevitable and probably should be taught in school. Now, as a new college admissions season kicks into gear, many prospective applicants are wondering: can AI write my personal essay?

Large Language Models (LLMs) have exploded a new heatwave of AI for their ability to engage end-users in human-level conversations with detailed and articulate answers across many knowledge domains. In response to their fast adoption in many industrial applications, this survey concerns their safety and trustworthiness. First, we review known vulnerabilities and limitations of the LLMs, categorising them into inherent issues, attacks, and unintended bugs. Then, we consider if and how the Verification and Validation (V&V) techniques, which have been widely developed for traditional software and deep learning models such as convolutional neural networks as independent processes to check the alignment of their implementations against the specifications, can be integrated and further extended throughout the lifecycle of the LLMs to provide rigorous analysis to the safety and trustworthiness of LLMs and their applications. Specifically, we consider four complementary techniques: falsification and evaluation, verification, runtime monitoring, and regulations and ethical use. In total, 370+ references are considered to support the quick understanding of the safety and trustworthiness issues from the perspective of V&V. While intensive research has been conducted to identify the safety and trustworthiness issues, rigorous yet practical methods are called for to ensure the alignment of LLMs with safety and trustworthiness requirements.

Current AI regulations require discarding sensitive features (e.g., gender, race, religion) in the algorithm's decision-making process to prevent unfair outcomes. However, even without sensitive features in the training set, algorithms can persist in discrimination. Indeed, when sensitive features are omitted (fairness under unawareness), they could be inferred through non-linear relations with the so called proxy features. In this work, we propose a way to reveal the potential hidden bias of a machine learning model that can persist even when sensitive features are discarded. This study shows that it is possible to unveil whether the black-box predictor is still biased by exploiting counterfactual reasoning. In detail, when the predictor provides a negative classification outcome, our approach first builds counterfactual examples for a discriminated user category to obtain a positive outcome. Then, the same counterfactual samples feed an external classifier (that targets a sensitive feature) that reveals whether the modifications to the user characteristics needed for a positive outcome moved the individual to the non-discriminated group. When this occurs, it could be a warning sign for discriminatory behavior in the decision process. Furthermore, we leverage the deviation of counterfactuals from the original sample to determine which features are proxies of specific sensitive information. Our experiments show that, even if the model is trained without sensitive features, it often suffers discriminatory biases.

In May, the CEOs of three of the most prominent AI labs--OpenAI, Google DeepMind, and Anthropic--signed a statement that warned AI could be as risky to humanity as pandemics and nuclear war. To prevent disaster, many AI companies and researchers are arguing for restrictions on who can access the most powerful AI models and who can develop them in the first place. They worry that bad actors could use AI models to create large amounts of disinformation that could alter the outcomes of elections, and that in the future, more powerful AI models could help launch cyberattacks or create bioweapons. But not all AI companies agree. On Thursday, Meta released Code Llama, a family of AI models built on top of Llama 2, Meta's flagship large language model, with extra training to make them particularly useful for coding tasks.

A TikToker warned of a growing trend involving child predators who use artificial intelligence to turn photos and videos of kids into explicit content. Posting imagery of children on social media can invite "digital kidnappers" to steal their likeness and use them in exploitative AI-generated videos, Alex Hoffman said in a viral TikTok video. "Digital kidnapping is when somebody steals the photos of your minor from the internet, usually a social media platform, and either pretends to be the child or pretends to be the child's parents," she said. "Oftentimes digital kidnappers will take normal photos of a child on the internet and alter them to look explicit or show the child doing something inappropriate." "Digital kidnappers can also take photos of a child and make them into an inappropriate video using AI materials," said Hoffman, a law student who has worked with the government investigating online sex crimes against children.

Sam Altman, the CEO of artificial intelligence lab OpenAI, told a Senate panel he welcomes federal regulation on the technology "to mitigate" its risks. The panic to regulate artificial intelligence (AI) came almost immediately after last fall's release of ChatGPT popularized the technology with the public. Some industry insiders themselves called for a pause on development, highlighting that expertise in a field doesn't translate into proficiency in the perils of regulation. That appeal was followed by a White House AI Bill of Rights and an educational effort by Senate Majority Leader Chuck Schumer, D-N.Y. Fears about AI include job displacement, data security and privacy, misinformation, autonomous defense systems mistakes, discrimination and bias, and an existential threat to humanity itself. It's imperative to prove actual market failure before regulating and to make sure the costs of doing so don't outweigh the benefits.

This paper presents a novel approach named Persona-Grouping-Intelligence (PGI), which has been crafted to tackle the challenges posed by GPT models when applied to real-world business issues. PGI leverages the inherent capabilities of the GPT model to comprehend intricate language structures and generate responses that are contextually relevant. The experiment occurred in a business scenario where human intelligence was being underutilized due to less optimized business processes. The primary objective of this approach is to leverage GPT models to reduce the workload on humans in tasks that are extensive, monotonous, and repetitive. Instead, the focus is redirected toward decision-making activities. Remarkably, the experiment yielded an accuracy rate of 93.81% in validating 4,000 responses generated by the model, underscoring the effectiveness of the PGI strategies. Effectively addressing the issue of underutilized human intelligence, this paradigm shift aligns business environments with dynamic machine intelligence, enabling them to navigate the intricacies of real-world challenges. This approach facilitates the practical utilization of these models to tackle actual problems. The methodology offers an opportunity to reshape the fundamental structure of business processes by seamlessly integrating human decision-making with adaptable machine intelligence. Consequently, this optimization enhances operational efficiency and elevates strategic decision-making across diverse business contexts.

Chat Generative Pre-trained Transformer (ChatGPT) has gained significant interest and attention since its launch in November 2022. It has shown impressive performance in various domains, including passing exams and creative writing. However, challenges and concerns related to biases and trust persist. In this work, we present a comprehensive review of over 100 Scopus-indexed publications on ChatGPT, aiming to provide a taxonomy of ChatGPT research and explore its applications. We critically analyze the existing literature, identifying common approaches employed in the studies. Additionally, we investigate diverse application areas where ChatGPT has found utility, such as healthcare, marketing and financial services, software engineering, academic and scientific writing, research and education, environmental science, and natural language processing. Through examining these applications, we gain valuable insights into the potential of ChatGPT in addressing real-world challenges. We also discuss crucial issues related to ChatGPT, including biases and trustworthiness, emphasizing the need for further research and development in these areas. Furthermore, we identify potential future directions for ChatGPT research, proposing solutions to current challenges and speculating on expected advancements. By fully leveraging the capabilities of ChatGPT, we can unlock its potential across various domains, leading to advancements in conversational AI and transformative impacts in society.

While analyzing scanned documents, handwritten text can overlap with printed text. This overlap causes difficulties during the optical character recognition (OCR) and digitization process of documents, and subsequently, hurts downstream NLP tasks. Prior research either focuses solely on the binary classification of handwritten text or performs a three-class segmentation of the document, i.e., recognition of handwritten, printed, and background pixels. This approach results in the assignment of overlapping handwritten and printed pixels to only one of the classes, and thus, they are not accounted for in the other class. Thus, in this research, we develop novel approaches to address the challenges of handwritten and printed text segmentation. Our objective is to recover text from different classes in their entirety, especially enhancing the segmentation performance on overlapping sections. To support this task, we introduce a new dataset, SignaTR6K, collected from real legal documents, as well as a new model architecture for the handwritten and printed text segmentation task. Our best configuration outperforms prior work on two different datasets by 17.9% and 7.3% on IoU scores. The SignaTR6K dataset is accessible for download via the following link: https://forms.office.com/r/2a5RDg7cAY.

Transformer language models have received widespread public attention, yet their generated text is often surprising even to NLP researchers. In this survey, we discuss over 250 recent studies of English language model behavior before task-specific fine-tuning. Language models possess basic capabilities in syntax, semantics, pragmatics, world knowledge, and reasoning, but these capabilities are sensitive to specific inputs and surface features. Despite dramatic increases in generated text quality as models scale to hundreds of billions of parameters, the models are still prone to unfactual responses, commonsense errors, memorized text, and social biases. Many of these weaknesses can be framed as over-generalizations or under-generalizations of learned patterns in text. We synthesize recent results to highlight what is currently known about large language model capabilities, thus providing a resource for applied work and for research in adjacent fields that use language models.