Pretrained language models underpin several AI applications, but their high computational cost for training limits accessibility. Initiatives such as BLOOM and StarCoder aim to democratize access to pretrained models for collaborative community development. However, such existing models face challenges: limited multilingual capabilities, continual pretraining causing catastrophic forgetting, whereas pretraining from scratch is computationally expensive, and compliance with AI safety and development laws. This paper presents Aurora-M, a 15B parameter multilingual open-source model trained on English, Finnish, Hindi, Japanese, Vietnamese, and code. Continually pretrained from StarCoderPlus on 435 billion additional tokens, Aurora-M surpasses 2 trillion tokens in total training token count. It is the first open-source multilingual model fine-tuned on human-reviewed safety instructions, thus aligning its development not only with conventional red-teaming considerations, but also with the specific concerns articulated in the Biden-Harris Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence. Aurora-M is rigorously evaluated across various tasks and languages, demonstrating robustness against catastrophic forgetting and outperforming alternatives in multilingual settings, particularly in safety evaluations. To promote responsible open-source LLM development, Aurora-M and its variants are released at https://huggingface.co/collections/aurora-m/aurora-m-models-65fdfdff62471e09812f5407 .

Recent generative AI systems have demonstrated more advanced persuasive capabilities and are increasingly permeating areas of life where they can influence decision-making. Generative AI presents a new risk profile of persuasion due the opportunity for reciprocal exchange and prolonged interactions. This has led to growing concerns about harms from AI persuasion and how they can be mitigated, highlighting the need for a systematic study of AI persuasion. The current definitions of AI persuasion are unclear and related harms are insufficiently studied. Existing harm mitigation approaches prioritise harms from the outcome of persuasion over harms from the process of persuasion. In this paper, we lay the groundwork for the systematic study of AI persuasion. We first put forward definitions of persuasive generative AI. We distinguish between rationally persuasive generative AI, which relies on providing relevant facts, sound reasoning, or other forms of trustworthy evidence, and manipulative generative AI, which relies on taking advantage of cognitive biases and heuristics or misrepresenting information. We also put forward a map of harms from AI persuasion, including definitions and examples of economic, physical, environmental, psychological, sociocultural, political, privacy, and autonomy harm. We then introduce a map of mechanisms that contribute to harmful persuasion. Lastly, we provide an overview of approaches that can be used to mitigate against process harms of persuasion, including prompt engineering for manipulation classification and red teaming. Future work will operationalise these mitigations and study the interaction between different types of mechanisms of persuasion.

With recent legislation on the right to be forgotten, machine unlearning has emerged as a crucial research area. It facilitates the removal of a user's data from federated trained machine learning models without the necessity for retraining from scratch. However, current machine unlearning algorithms are confronted with challenges of efficiency and validity. To address the above issues, we propose a new framework, named Goldfish. It comprises four modules: basic model, loss function, optimization, and extension. To address the challenge of low validity in existing machine unlearning algorithms, we propose a novel loss function. It takes into account the loss arising from the discrepancy between predictions and actual labels in the remaining dataset. Simultaneously, it takes into consideration the bias of predicted results on the removed dataset. Moreover, it accounts for the confidence level of predicted results. Additionally, to enhance efficiency, we adopt knowledge a distillation technique in the basic model and introduce an optimization module that encompasses the early termination mechanism guided by empirical risk and the data partition mechanism. Furthermore, to bolster the robustness of the aggregated model, we propose an extension module that incorporates a mechanism using adaptive distillation temperature to address the heterogeneity of user local data and a mechanism using adaptive weight to handle the variety in the quality of uploaded models. Finally, we conduct comprehensive experiments to illustrate the effectiveness of proposed approach.

Artificial neural networks are often interpreted as abstract models of biological neuronal networks, but they are typically trained using the biologically unrealistic backpropagation algorithm and its variants. Predictive coding has been proposed as a potentially more biologically realistic alternative to backpropagation for training neural networks. This manuscript reviews and extends recent work on the mathematical relationship between predictive coding and backpropagation for training feedforward artificial neural networks on supervised learning tasks. Implications of these results for the interpretation of predictive coding and deep neural networks as models of biological learning are discussed along with a repository of functions, Torch2PC, for performing predictive coding with PyTorch neural network models.

Survival regression aims to predict the time when an event of interest will take place, typically a death or a failure. A fully parametric method [18] is proposed to estimate the survival function as a mixture of individual parametric distributions in the presence of censoring. In this paper, We present a novel method to predict the survival time by better clustering the survival data and combine primitive distributions. We propose two variants of variational auto-encoder (VAE), discrete and continuous, to generate the latent variables for clustering input covariates. The model is trained end to end by jointly optimizing the VAE loss and regression loss. Thorough experiments on dataset SUPPORT and FLCHAIN show that our method can effectively improve the clustering result and reach competitive scores with previous methods. We demonstrate the superior result of our model prediction in the long-term. Our code is available at https://github.com/

Paedophiles are being urged to use artificial intelligence to create nude images of children to extort more extreme material from them, according to a child abuse charity. The Internet Watch Foundation (IWF) said a manual found on the dark web contained a section encouraging criminals to use "nudifying" tools to remove clothing from underwear shots sent by a child. The manipulated image could then be used against the child to blackmail them into sending more graphic content, the IWF said. "This is the first evidence we have seen that perpetrators are advising and encouraging each other to use AI technology for these ends," said the IWF. The charity, which finds and removes child sexual abuse material online, warned last year of a rise in sextortion cases, where victims are manipulated into sending graphic images of themselves and are then threatened with the release of those images unless they hand over money.

Child abusers are creating AI-generated "deepfakes" of their targets in order to blackmail them into filming their own abuse, beginning a cycle of sextortion that can last for years. Creating simulated child abuse imagery is illegal in the UK, and Labour and the Conservatives have aligned on the desire to ban all explicit AI-generated images of real people. But there is little global agreement on how the technology should be policed. Worse, no matter how strongly governments take action, the creation of more images will always be a press of a button away – explicit imagery is built into the foundations of AI image generation. In December, researchers at Stanford University made a disturbing discovery: buried among the billions of images making up one of the largest training sets for AI image generators was hundreds, maybe thousands, of instances of child sexual abuse material (CSAM).

Senate Intelligence Committee member Marco Rubio, R-Fla., tells'Hannity' the idea of citizenship is in danger. The Supreme Court announced Monday that it will hear a challenge to the Biden administration's regulation on so-called "ghost guns" next term. The rule in question was issued in 2022 by the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) to regulate "buy build shoot" kits that are available online or in stores that allow any individual to assemble a working firearm without a background check or the usual serial numbers required by the federal government. The Fifth Circuit late last year struck down the rule, but the Justice Department appealed to the Supreme Court. The DOJ argued that the Gun Control Act of 1968 permits the rule because it defines a "firearm" to include "any weapon…which will or is designed to or may readily be converted to expel a projectile by the action of an explosive," as well as "the frame or receiver of any such weapon."

But just 5 to 8 percent of those reports ever lead to arrests, the report said, due to a shortage of funding and resources, legal constraints, and a cascade of shortcomings in the process for reporting, prioritizing and investigating them. If those limitations aren't addressed soon, the authors warn, the system could become unworkable as the latest AI image generators unleash a deluge of sexual imagery of virtual children that is increasingly "indistinguishable from real photos of children."

Work is fundamental to societal prosperity and mental health, providing financial security, identity, purpose, and social integration. The emergence of generative artificial intelligence (AI) has catalysed debate on job displacement. Some argue that many new jobs and industries will emerge to offset the displacement, while others foresee a widespread decoupling of economic productivity from human input threatening jobs on an unprecedented scale. This study explores the conditions under which both may be true and examines the potential for a self-reinforcing cycle of recessionary pressures that would necessitate sustained government intervention to maintain job security and economic stability. A system dynamics model was developed to undertake ex ante analysis of the effect of AI-capital deepening on labour underutilisation and demand in the economy. Results indicate that even a moderate increase in the AI-capital-to-labour ratio could increase labour underutilisation to double its current level, decrease per capita disposable income by 26% (95% interval, 20.6% - 31.8%), and decrease the consumption index by 21% (95% interval, 13.6% - 28.3%) by mid-2050. To prevent a reduction in per capita disposable income due to the estimated increase in underutilization, at least a 10.8-fold increase in the new job creation rate would be necessary. Results demonstrate the feasibility of an AI-capital- to-labour ratio threshold beyond which even high rates of new job creation cannot prevent declines in consumption. The precise threshold will vary across economies, emphasizing the urgent need for empirical research tailored to specific contexts. This study underscores the need for governments, civic organisations, and business to work together to ensure a smooth transition to an AI- dominated economy to safeguard the Mental Wealth of nations.