We introduce the Concept Bottleneck Large Language Model (CB-LLM), a pioneering approach to creating inherently interpretable Large Language Models (LLMs). Unlike traditional black-box LLMs that rely on post-hoc interpretation methods with limited neuron function insights, CB-LLM sets a new standard with its built-in interpretability, scalability, and ability to provide clear, accurate explanations. We investigate two essential tasks in the NLP domain: text classification and text generation. In text classification, CB-LLM narrows the performance gap with traditional black-box models and provides clear interpretability. In text generation, we show how interpretable neurons in CB-LLM can be used for concept detection and steering text generation. Our CB-LLMs enable greater interaction between humans and LLMs across a variety of tasks -- a feature notably absent in existing LLMs. Large Language Models (LLMs) have become instrumental in advancing Natural Language Processing (NLP) tasks.

LOS ANGELES/WASHINGTON, Nov 15 (Thomson Reuters Foundation) - When the sheriff in Suffolk County, New York, requested $700,000 from the U.S. government for an artificial intelligence system to eavesdrop on prison phone conversations, his office called it a key tool in fighting gang-related and violent crime. But the county jail ended up listening to calls involving a much wider range of subjects - scanning as many as 600,000 minutes per month, according to public records from the county obtained by the Thomson Reuters Foundation. Beginning in 2019, Suffolk County was an early pilot site for the Verus AI-scanning system sold by California-based LEO Technologies, which uses Amazon speech-to-text technology to transcribe phone calls flagged by key word searches. The company and law enforcement officials say it is a crucial tool to keep prisons and jails safe, and fight crime, but critics say such systems trample the privacy rights of prisoners and other people, like family members, on the outside. "The ability to surveil and listen at scale in this rapid way - it is incredibly scary and chilling," said Julie Mao, deputy director at Just Futures Law, an immigration legal group.

When the sheriff in Suffolk County, New York, requested $700,000 from the U.S. government for an artificial intelligence system to eavesdrop on prison phone conversations, his office called it a key tool in fighting gang-related and violent crime. But the county jail ended up listening to calls involving a much wider range of subjects -- scanning as many as 600,000 minutes per month, according to public records from the county obtained by the Thomson Reuters Foundation. Beginning in 2019, Suffolk County was an early pilot site for the Verus AI-scanning system sold by California-based LEO Technologies, which uses Amazon speech-to-text technology to transcribe phone calls flagged by keyword searches. The company and law enforcement officials say it is a crucial tool to keep prisons and jails safe, and to fight crime, but critics say such systems trample the privacy rights of prisoners and other people, like family members, on the outside. "T he ability to surveil and listen at scale in this rapid way -- it is incredibly scary and chilling," said Julie Mao, deputy director at Just Futures Law, an immigration legal group.

Corruption continues to be one of the biggest societal challenges of our time. New hope is placed in Artificial Intelligence (AI) to serve as an unbiased anti-corruption agent. Ever more available (open) government data paired with unprecedented performance of such algorithms render AI the next frontier in anti-corruption. Summarizing existing efforts to use AI-based anti-corruption tools (AI-ACT), we introduce a conceptual framework to advance research and policy. It outlines why AI presents a unique tool for top-down and bottom-up anti-corruption approaches. For both approaches, we outline in detail how AI-ACT present different potentials and pitfalls for (a) input data, (b) algorithmic design, and (c) institutional implementation. Finally, we venture a look into the future and flesh out key questions that need to be addressed to develop AI-ACT while considering citizens' views, hence putting "society in the loop".

Each year, around 6 million car accidents occur in the U.S. on average. Road safety features (e.g., concrete barriers, metal crash barriers, rumble strips) play an important role in preventing or mitigating vehicle crashes. Accurate maps of road safety features is an important component of safety management systems for federal or state transportation agencies, helping traffic engineers identify locations to invest on safety infrastructure. In current practice, mapping road safety features is largely done manually (e.g., observations on the road or visual interpretation of streetview imagery), which is both expensive and time consuming. In this paper, we propose a deep learning approach to automatically map road safety features from streetview imagery. Unlike existing Convolutional Neural Networks (CNNs) that classify each image individually, we propose to further add Recurrent Neural Network (Long Short Term Memory) to capture geographic context of images (spatial autocorrelation effect along linear road network paths). Evaluations on real world streetview imagery show that our proposed model outperforms several baseline methods.