Meta is under pressure to explain why it cancelled a major contract with a company it was using to train AI, shortly after some of its Kenya-based workers alleged they had to view graphic content captured by Meta smart glasses. In February, workers at the company, Sama, told two Swedish newspapers they had witnessed glasses users going to the toilet and having sex . Less than two months later, Meta ended its contract with Sama, which Sama said would result in 1,108 workers being made redundant. Meta says it's because Sama did not meet its standards, a criticism Sama rejects. A Kenyan workers' organisation alleges Meta's decision was caused by the staff speaking out.

The Utah court of appeals has sanctioned a lawyer after he was discovered to have used ChatGPT for a filing he made in which he referenced a nonexistent court case. Earlier this week, the Utah court of appeals made the decision to sanction Richard Bednar over claims that he filed a brief which included false citations. According to court documents reviewed by ABC4, Bednar and Douglas Durbano, another Utah-based lawyer who was serving as the petitioner's counsel, filed a "timely petition for interlocutory appeal". Upon reviewing the brief which was written by a law clerk, the respondent's counsel found several false citations of cases. "It appears that at least some portions of the Petition may be AI-generated, including citations and even quotations to at least one case that does not appear to exist in any legal database (and could only be found in ChatGPT and references to cases that are wholly unrelated to the referenced subject matter," the respondent's counsel said in documents reviewed by ABC4.

In this paper, we present PARAMANU-AYN, a language model based exclusively on case documents of the Supreme Court of India, the Constitution of India, and the Indian Penal Code. The novel Auto Regressive (AR) decoder based model is pretrained from scratch at a context size of 8192. We evaluated our pretrained legal model on perplexity metrics. We also instruction-tuned our pretrained model on a set of 10,763 instructions covering various legal tasks such as legal reasoning, judgement explanation, legal clause generation, legal drafting, legal contract drafting, case summarization, constitutional question-answering, etc. We also evaluated the responses of prompts for instruction-tuned models by GPT-3.5-Turbo on clarity, relevance, completeness, and legal reasoning metrics in a scale of 10. Our model can be run on CPU and achieved 42.46 tokens/sec CPU inference speed. We found that our models, despite not being pretrained on legal books, various legal contracts, and legal documents, were able to learn the domain knowledge required for drafting various legal contracts and legal clauses, and generalize to draft legal contracts and legal clauses with limited instruction tuning. Hence, we conclude that for a strong domain-specialized generative language model (such as legal), very large amounts of data are not required to develop models from scratch. We believe that this work is the first attempt to make a dedicated generative legal language model from scratch for Indian Supreme Court jurisdiction or in legal NLP overall. We plan to release our Paramanu-Ayn model at https://www.bharatgpts.com.

The images pop up in Mophat Okinyi's mind when he's alone, or when he's about to sleep. Okinyi, a former content moderator for Open AI's ChatGPT in Nairobi, Kenya, is one of four people in that role who have filed a petition to the Kenyan government calling for an investigation into what they describe as exploitative conditions for contractors reviewing the content that powers artificial intelligence programs. "It has really damaged my mental health," said Okinyi. The 27-year-old said he would would view up to 700 text passages a day, many depicting graphic sexual violence. He recalls he started avoiding people after having read texts about rapists and found himself projecting paranoid narratives on to people around him.

In 1955, science fiction writer Isaac Asimov published a short story about an experiment in "electronic democracy," in which a single citizen, selected to represent an entire population, responded to questions generated by a computer named Multivac. The machine took this data and calculated the results of an election that therefore never needed to happen. Asimov's story was set in Bloomington, Indiana, but today an approximation of Multivac is being built in China. For any authoritarian regime, "there is a basic problem for the center of figuring out what's going on at lower levels and across society," says Deborah Seligsohn, a political scientist and China expert at Villanova University in Philadelphia. How do you effectively govern a country that's home to one in five people on the planet, with an increasingly complex economy and society, if you don't allow public debate, civil activism, and electoral feedback?

We explore the idea that authoring a piece of text is an act of maximizing one's expected utility.To make this idea concrete, we consider the societally important decisions of the Supreme Court of the United States.Extensive past work in quantitative political science provides a framework for empirically modeling the decisions of justices and how they relate to text.We incorporate into such a model texts authored by amici curiae (``friends of the court'' separate from the litigants) who seek to weigh in on the decision, then explicitly model their goals in a random utility model.We demonstrate the benefits of this approach in improved vote prediction and the ability to perform counterfactual analysis.

We explore the idea that authoring a piece of text is an act of maximizing one's expected utility. To make this idea concrete, we consider the societally important decisions of the Supreme Court of the United States. Extensive past work in quantitative political science provides a framework for empirically modeling the decisions of justices and how they relate to text. We incorporate into such a model texts authored by amici curiae ("friends of the court" separate from the litigants) who seek to weigh in on the decision, then explicitly model their goals in a random utility model. We demonstrate the benefits of this approach in improved vote prediction and the ability to perform counterfactual analysis.