We investigate whether an LLM can successfully perform financial statement analysis in a way similar to a professional human analyst. We provide standardized and anonymous financial statements to GPT4 and instruct the model to analyze them to determine the direction of future earnings. Even without any narrative or industry-specific information, the LLM outperforms financial analysts in its ability to predict earnings changes. The LLM exhibits a relative advantage over human analysts in situations when the analysts tend to struggle. Furthermore, we find that the prediction accuracy of the LLM is on par with the performance of a narrowly trained state-of-the-art ML model. LLM prediction does not stem from its training memory. Instead, we find that the LLM generates useful narrative insights about a company's future performance. Lastly, our trading strategies based on GPT's predictions yield a higher Sharpe ratio and alphas than strategies based on other models. Taken together, our results suggest that LLMs may take a central role in decision-making.

Federated learning has become a significant approach for training machine learning models using decentralized data without necessitating the sharing of this data. Recently, the incorporation of generative artificial intelligence (AI) methods has provided new possibilities for improving privacy, augmenting data, and customizing models. This research explores potential integrations of generative AI in federated learning, revealing various opportunities to enhance privacy, data efficiency, and model performance. It particularly emphasizes the importance of generative models like generative adversarial networks (GANs) and variational autoencoders (VAEs) in creating synthetic data that replicates the distribution of real data. Generating synthetic data helps federated learning address challenges related to limited data availability and supports robust model development. Additionally, we examine various applications of generative AI in federated learning that enable more personalized solutions.

Synthetic data has become an important tool in the fine-tuning of language models to follow instructions and solve complex problems. Nevertheless, the majority of open data to date is often lacking multi-turn data and collected on closed models, limiting progress on advancing open fine-tuning methods. We introduce Self Directed Synthetic Dialogues (SDSD), an experimental dataset consisting of guided conversations of language models talking to themselves. The dataset consists of multi-turn conversations generated with DBRX, Llama 2 70B, and Mistral Large, all instructed to follow a conversation plan generated prior to the conversation. We also explore including principles from Constitutional AI and other related works to create synthetic preference data via revisions to the final conversation turn. We hope this work encourages further exploration in multi-turn data and the use of open models for expanding the impact of synthetic data.

As the reality TV sector increasingly has to deal with the good and bad impacts of AI, lawyer John Delaney says there are growing legal and regulatory issues. "For example, AI could be used to suggest scenarios or storylines, to edit episodes and to anticipate and assess audience reactions to in-show developments," says Mr Delaney, who is a partner at commercial law firm Perkins Coie, and who advises companies on AI and other technology issues. "However, production companies will need to consider to what extent the new Writers Guild of America agreement [to strictly restrict the use of AI] might limit their ability to use AI in connection with their reality TV programs." He adds that away from making the shows a growing issue that reality TV producers and contestants are facing is a proliferation of unauthorized, AI-generated images and videos. Mr Delaney points to generative AI tools such as chatbot ChatGPT being used to create new content from reality TV footage.

The US Senate unanimously passed a bill on Tuesday designed to hold accountable those who make or share deepfake porn. The Disrupt Explicit Forged Images and Non-Consensual Edits Act (DEFIANCE Act) would allow victims to sue those who create, share or possess AI-generated sexual images or videos using their likeness. The issue took root in the public consciousness after the infamous Taylor Swift deepfake that circulated among online lowlifes early this year. The bill would let victims sue for up to 150,000 in damages. That number grows to 250,000 if it's related to attempted sexual assault, stalking or harassment.

When police departments first started buying and deploying bodycams in the wake of the police killing of Michael Brown in Ferguson, Missouri, a decade ago, activists hoped it would bring about real change. Years later, despite what's become a multibillion-dollar market for these devices, the tech is far from a panacea. Most footage they generate goes unwatched. And if they do finally provide video to the public, it usually doesn't tell the complete story. A handful of AI startups see this problem as an opportunity to create what are essentially bodycam-to-text programs for different players in the legal system, mining this footage for misdeeds. But like the bodycams themselves, the technology still faces procedural, legal, and cultural barriers to success.

You wake up on Election Day and unlock your phone to a shaky video of your state capitol. In other clips posted alongside it, gunshots ring out in the distance. You think to yourself: Maybe better to skip the polling booth today. Only later do you learn that the videos were AI forgeries. A friend calls you, distraught.

Large Language Models (LLMs) have been shown to perform well for many downstream tasks. Transfer learning can enable LLMs to acquire skills that were not targeted during pre-training. In financial contexts, LLMs can sometimes beat well-established benchmarks. This paper investigates how well LLMs perform in the task of forecasting corporate credit ratings. We show that while LLMs are very good at encoding textual information, traditional methods are still very competitive when it comes to encoding numeric and multimodal data. For our task, current LLMs perform worse than a more traditional XGBoost architecture that combines fundamental and macroeconomic data with high-density text-based embedding features.

Human-produced emissions are growing at an alarming rate, causing already observable changes in the climate and environment in general. Each year global carbon dioxide emissions hit a new record, and it is reported that 0.5% of total US greenhouse gas emissions are attributed to data centres as of 2021. The release of ChatGPT in late 2022 sparked social interest in Large Language Models (LLMs), the new generation of Language Models with a large number of parameters and trained on massive amounts of data. Currently, numerous companies are releasing products featuring various LLMs, with many more models in development and awaiting release. Deep Learning research is a competitive field, with only models that reach top performance attracting attention and being utilized. Hence, achieving better accuracy and results is often the first priority, while the model's efficiency and the environmental impact of the study are neglected. However, LLMs demand substantial computational resources and are very costly to train, both financially and environmentally. It becomes essential to raise awareness and promote conscious decisions about algorithmic and hardware choices. Providing information on training time, the approximate carbon dioxide emissions and power consumption would assist future studies in making necessary adjustments and determining the compatibility of available computational resources with model requirements. In this study, we infused T5 LLM with external knowledge and fine-tuned the model for Question-Answering task. Furthermore, we calculated and reported the approximate environmental impact for both steps. The findings demonstrate that the smaller models may not always be sustainable options, and increased training does not always imply better performance. The most optimal outcome is achieved by carefully considering both performance and efficiency factors.

Generative AI holds the promise of enabling a range of sought-after capabilities and revolutionizing workflows in various consumer and enterprise verticals. However, putting a model in production involves much more than just generating an output. It involves ensuring the model is reliable, safe, performant and also adheres to the policy of operation in a particular domain. Guardrails as a necessity for models has evolved around the need to enforce appropriate behavior of models, especially when they are in production. In this paper, we use education as a use case, given its stringent requirements of the appropriateness of content in the domain, to demonstrate how a guardrail model can be trained and deployed in production. Specifically, we describe our experience in building a production-grade guardrail model for a K-12 educational platform. We begin by formulating the requirements for deployment to this sensitive domain. We then describe the training and benchmarking of our domain-specific guardrail model, which outperforms competing open- and closed- instruction-tuned models of similar and larger size, on proprietary education-related benchmarks and public benchmarks related to general aspects of safety. Finally, we detail the choices we made on architecture and the optimizations for deploying this service in production; these range across the stack from the hardware infrastructure to the serving layer to language model inference optimizations. We hope this paper will be instructive to other practitioners looking to create production-grade domain-specific services based on generative AI and large language models.