In some ways the last two days have marked a real move forward in AI regulation. There is now an explicit consensus to take on an international shared responsibility. There are plans for further meetings in South Korea and France over the next year to continue the conversation surrounding risks posed by AI. However, critics have said that the focus on the more hypothetical aspects of existential threats as opposed to the more current, pressing problems – such as insurance companies using AI to determine your risk or AI's potential to take people's jobs – dilutes the impact of the summit. This is an edited extract from the First Edition newsletter – sign up here for our free daily email, taking you through the top stories and what they mean.

Canopy CMO Yaron Litwin discusses how criminals are using deepfake technology to blackmail teens and generate child pornography. Parents of girls at a New Jersey high school said their daughters were humiliated after they learned fake pornographic images of themselves generated with the use of Artificial Intelligence (AI), were circulated among classmates. Female students at Westfield High, NJ, learned about the images after observing sophomore boys whispering and acting "weird" on October 16, the Wall Street Journal reported. A few days later, one boy confessed to some of the girls that at least one student had used real photos, found on social media, to generate fake nude images of the female classmates through an AI website. The student then reportedly shared these images with other male students in a group text.

Most applications of Artificial Intelligence (AI) are designed for a confined and specific task. However, there are many scenarios that call for a more general AI, capable of solving a wide array of tasks without being specifically designed for them. The term General-Purpose Artificial Intelligence Systems (GPAIS) has been defined to refer to these AI systems. To date, the possibility of an Artificial General Intelligence, powerful enough to perform any intellectual task as if it were human, or even improve it, has remained an aspiration, fiction, and considered a risk for our society. Whilst we might still be far from achieving that, GPAIS is a reality and sitting at the forefront of AI research. This work discusses existing definitions for GPAIS and proposes a new definition that allows for a gradual differentiation among types of GPAIS according to their properties and limitations. We distinguish between closed-world and open-world GPAIS, characterising their degree of autonomy and ability based on several factors such as adaptation to new tasks, competence in domains not intentionally trained for, ability to learn from few data, or proactive acknowledgment of their own limitations. We propose a taxonomy of approaches to realise GPAIS, describing research trends such as the use of AI techniques to improve another AI (AI-powered AI) or (single) foundation models. As a prime example, we delve into GenAI, aligning them with the concepts presented in the taxonomy. We explore multi-modality, which involves fusing various types of data sources to expand the capabilities of GPAIS. Through the proposed definition and taxonomy, our aim is to facilitate research collaboration across different areas that are tackling general purpose tasks, as they share many common aspects. Finally, we discuss the state of GPAIS, prospects, societal implications, and the need for regulation and governance.

The development and regulation of multi-purpose, large "foundation models" of AI seems to have reached a critical stage, with major investments and new applications announced every other day. Some experts are calling for a moratorium on the training of AI systems more powerful than GPT-4. Legislators globally compete to set the blueprint for a new regulatory regime. This paper analyses the most advanced legal proposal, the European Union's AI Act currently in the stage of final "trilogue" negotiations between the EU institutions. This legislation will likely have extra-territorial implications, sometimes called "the Brussels effect". It also constitutes a radical departure from conventional information and communications technology policy by regulating AI ex-ante through a risk-based approach that seeks to prevent certain harmful outcomes based on product safety principles. We offer a review and critique, specifically discussing the AI Act's problematic obligations regarding data quality and human oversight. Our proposal is to take liability seriously as the key regulatory mechanism. This signals to industry that if a breach of law occurs, firms are required to know in particular what their inputs were and how to retrain the system to remedy the breach. Moreover, we suggest differentiating between endogenous and exogenous sources of potential harm, which can be mitigated by carefully allocating liability between developers and deployers of AI technology.

The European debt purchase market as measured by the total book value of purchased debt approached 25bn euros in 2020 and it was growing at double-digit rates. This is an example of how big the debt collection and debt purchase industry has grown and the important impact it has in the financial sector. However, in order to ensure an adequate return during the debt collection process, a good estimation of the propensity to pay and/or the expected cashflow is crucial. These estimations can be employed, for instance, to create different strategies during the amicable collection to maximize quality standards and revenues. And not only that, but also to prioritize the cases in which a legal process is necessary when debtors are unreachable for an amicable negotiation. This work offers a solution for these estimations. Specifically, a new machine learning modelling pipeline is presented showing how outperforms current strategies employed in the sector. The solution contains a pre-processing pipeline and a model selector based on the best model calibration. Performance is validated with real historical data of the debt industry.

While a substantial amount of work has recently been devoted to enhance the performance of computational Authorship Identification (AId) systems, little to no attention has been paid to endowing AId systems with the ability to explain the reasons behind their predictions. This lacking substantially hinders the practical employment of AId methodologies, since the predictions returned by such systems are hardly useful unless they are supported with suitable explanations. In this paper, we explore the applicability of existing general-purpose eXplainable Artificial Intelligence (XAI) techniques to AId, with a special focus on explanations addressed to scholars working in cultural heritage. In particular, we assess the relative merits of three different types of XAI techniques (feature ranking, probing, factuals and counterfactual selection) on three different AId tasks (authorship attribution, authorship verification, same-authorship verification) by running experiments on real AId data. Our analysis shows that, while these techniques make important first steps towards explainable Authorship Identification, more work remains to be done in order to provide tools that can be profitably integrated in the workflows of scholars.

Identifying contextual integrity (CI) and governing knowledge commons (GKC) parameters in privacy policy texts can facilitate normative privacy analysis. However, GKC-CI annotation has heretofore required manual or crowdsourced effort. This paper demonstrates that high-accuracy GKC-CI parameter annotation of privacy policies can be performed automatically using large language models. We fine-tune 18 open-source and proprietary models on 21,588 GKC-CI annotations from 16 ground truth privacy policies. Our best-performing model (fine-tuned GPT-3.5 Turbo with prompt engineering) has an accuracy of 86%, exceeding the performance of prior crowdsourcing approaches despite the complexity of privacy policy texts and the nuance of the GKC-CI annotation task. We apply our best-performing model to privacy policies from 164 popular online services, demonstrating the effectiveness of scaling GKC-CI annotation for data exploration. We make all annotated policies as well as the training data and scripts needed to fine-tune our best-performing model publicly available for future research.

Online forums encourage the exchange and discussion of different stances on many topics. Not only do they provide an opportunity to present one's own arguments, but may also gather a broad cross-section of others' arguments. However, the resulting long discussions are difficult to overview. This paper presents a novel unsupervised approach using large language models (LLMs) to generating indicative summaries for long discussions that basically serve as tables of contents. Our approach first clusters argument sentences, generates cluster labels as abstractive summaries, and classifies the generated cluster labels into argumentation frames resulting in a two-level summary. Based on an extensively optimized prompt engineering approach, we evaluate 19~LLMs for generative cluster labeling and frame classification. To evaluate the usefulness of our indicative summaries, we conduct a purpose-driven user study via a new visual interface called Discussion Explorer: It shows that our proposed indicative summaries serve as a convenient navigation tool to explore long discussions.

Although large language models (LLMs) are widely deployed, the data used to train them is rarely disclosed. Given the incredible scale of this data, up to trillions of tokens, it is all but certain that it includes potentially problematic text such as copyrighted materials, personally identifiable information, and test data for widely reported reference benchmarks. However, we currently have no way to know which data of these types is included or in what proportions. In this paper, we study the pretraining data detection problem: given a piece of text and black-box access to an LLM without knowing the pretraining data, can we determine if the model was trained on the provided text? To facilitate this study, we introduce a dynamic benchmark WIKIMIA that uses data created before and after model training to support gold truth detection. We also introduce a new detection method Min-K% Prob based on a simple hypothesis: an unseen example is likely to contain a few outlier words with low probabilities under the LLM, while a seen example is less likely to have words with such low probabilities. Min-K% Prob can be applied without any knowledge about the pretraining corpus or any additional training, departing from previous detection methods that require training a reference model on data that is similar to the pretraining data. Moreover, our experiments demonstrate that Min-K% Prob achieves a 7.4% improvement on WIKIMIA over these previous methods. We apply Min-K% Prob to three real-world scenarios, copyrighted book detection, contaminated downstream example detection and privacy auditing of machine unlearning, and find it a consistently effective solution.

Recent discoveries have revealed that deep neural networks might behave in a biased manner in many real-world scenarios. For instance, deep networks trained on a large-scale face recognition dataset CelebA tend to predict blonde hair for females and black hair for males. Such biases not only jeopardize the robustness of models but also perpetuate and amplify social biases, which is especially concerning for automated decision-making processes in healthcare, recruitment, etc., as they could exacerbate unfair economic and social inequalities among different groups. Existing debiasing methods suffer from high costs in bias labeling or model re-training, while also exhibiting a deficiency in terms of elucidating the origins of biases within the model. To this respect, we propose a fast model debiasing framework (FMD) which offers an efficient approach to identify, evaluate and remove biases inherent in trained models. The FMD identifies biased attributes through an explicit counterfactual concept and quantifies the influence of data samples with influence functions. Moreover, we design a machine unlearning-based strategy to efficiently and effectively remove the bias in a trained model with a small counterfactual dataset. Experiments on the Colored MNIST, CelebA, and Adult Income datasets along with experiments with large language models demonstrate that our method achieves superior or competing accuracies compared with state-of-the-art methods while attaining significantly fewer biases and requiring much less debiasing cost. Notably, our method requires only a small external dataset and updating a minimal amount of model parameters, without the requirement of access to training data that may be too large or unavailable in practice.