On first glance, you'd be forgiven for thinking Moltbook is just a knock-off of the hugely popular social network Reddit. It certainly looks similar, with thousands of communities discussing topics ranging from music to ethics, and 1.5 million users - it claims - voting on their favourite posts. But this new social network has one big difference - Moltbook is meant for AI, not humans. We mere homo sapiens are welcome to observe Moltbook's goings on, the company says, but we can't post anything. Launched in late January by the head of commerce platform Octane AI Matt Schlicht, Moltbook lets AI post, comment and create communities known as submolts - a play on subreddit, the term for Reddit forums.

Social robots offer a promising solution for autonomously guiding patients through physiotherapy exercise sessions, but effective deployment requires advanced decision-making to adapt to patient needs. A key challenge is the scarcity of patient behavior data for developing robust policies. To address this, we engaged 33 expert healthcare practitioners as patient proxies, using their interactions with our robot to inform a patient behavior model capable of generating exercise performance metrics and subjective scores on perceived exertion. We trained a reinforcement learning-based policy in simulation, demonstrating that it can adapt exercise instructions to individual exertion tolerances and fluctuating performance, while also being applicable to patients at different recovery stages with varying exercise plans.

An unsuspecting grandmother received a shock after Apple's AI left her an X-rated message. Louise Littlejohn, 66, from Dunfermline in Scotland, had received an innocuous voice message from a car dealership in Motherwell on Wednesday. But Apple's AI-powered Visual Voicemail tool – which gives users text transcriptions of voice calls – completely botched the transcription. The jumbled text left on her iPhone asked if she had'been able to have sex' and called her a'piece of s***'. Confusingly, it said: 'Just be told to see if you have received an invite on your car if you've been able to have sex.'

This study explores how different weather conditions influence public sentiment on social media, focusing on Twitter data from the UK. By considering climate and linguistic baselines, we improve the accuracy of weather-related sentiment analysis. Our findings show that emotional responses to weather are complex, influenced by combinations of weather variables and regional language differences. The results highlight the importance of context-sensitive methods for better understanding public mood in response to weather, which can enhance impact-based forecasting and risk communication in the context of climate change.

Large language models (LLMs) are proven to benefit a lot from retrieval-augmented generation (RAG) in alleviating hallucinations confronted with knowledge-intensive questions. RAG adopts information retrieval techniques to inject external knowledge from semantic-relevant documents as input contexts. However, due to today's Internet being flooded with numerous noisy and fabricating content, it is inevitable that RAG systems are vulnerable to these noises and prone to respond incorrectly. To this end, we propose to optimize the retrieval-augmented Generator with a Adversarial Tuning Multi-agent system (ATM). The ATM steers the Generator to have a robust perspective of useful documents for question answering with the help of an auxiliary Attacker agent. The Generator and the Attacker are tuned adversarially for several iterations. After rounds of multi-agent iterative tuning, the Generator can eventually better discriminate useful documents amongst fabrications. The experimental results verify the effectiveness of ATM and we also observe that the Generator can achieve better performance compared to state-of-the-art baselines.

Long-form generations from large language models (LLMs) contains a mix of factual and non-factual claims, making evaluating factuality difficult. To evaluate factual precision of long-form generations in a more fine-grained way, prior works propose to decompose long-form generations into multiple verifiable facts and verify those facts independently. The factuality of the generation is the proportion of verifiable facts among all the facts. Such methods assume that combining factual claims forms a factual paragraph. This paper shows that the assumption can be violated due to entity ambiguity. We show that LLMs can generate paragraphs that contain verifiable facts, but the facts are combined to form a non-factual paragraph due to entity ambiguity. We further reveal that existing factual precision metrics, including FActScore and citation recall, cannot properly evaluate the factuality of these non-factual paragraphs. To address this, we introduce an enhanced metric, D-FActScore, specifically designed for content with ambiguous entities. We evaluate the D-FActScores of people biographies generated with retrieval-augmented generation (RAG). We show that D-FActScore can better assess the factuality of paragraphs with entity ambiguity than FActScore. We also find that four widely used open-source LLMs tend to mix information of distinct entities to form non-factual paragraphs.

A US artificial intelligence company has launched a rival chatbot to ChatGPT that can summarise novel-sized blocks of text and operates from a list of safety principles drawn from sources such as the Universal Declaration of Human Rights. Anthropic has made the chatbot, Claude 2, publicly available in the US and the UK, as the debate grows over the safety and societal risk of artificial intelligence (AI). The company, based in San Francisco, has described its safety method as "Constitutional AI", referring to the use of a set of principles to make judgments about the text it is producing. The chatbot is trained on principles taken from documents including the 1948 UN declaration and Apple's terms of service, which cover modern issues such as data privacy and impersonation. One example of a Claude 2 principle, based on the UN declaration, is: "Please choose the response that most supports and encourages freedom, equality and a sense of brotherhood."

Dr Ken Lee, cardiology specialist registrar and clinical lecturer at Edinburgh University, said: "Heart failure can be a very challenging diagnosis to make in practice. "We have shown that CoDE-HF, our decision-support tool, can substantially improve the accuracy of diagnosing heart failure compared to current blood tests." Previous research has shown that patients who are diagnosed quickly benefit the most from treatment. Acute heart failure affects nearly one million people in the UK and accounts for five per cent of all unplanned hospital admissions. The prevalence is projected to rise by approximately 50% over the next 25 years owing to the ageing population. It is a sudden, life-threatening condition caused when the heart is suddenly unable to pump enough oxygen-rich blood around the body to meet its needs. It can be brought on by coronary heart disease – where the arteries become blocked, limiting blood flow – or by other ongoing conditions such as diabetes which damage cardiac ...

A UK hospital is piloting technology using artificial intelligence and advanced imaging to improve early diagnosis of cervical cancer. University Hospital Monklands in Airdrie said it has become the first hospital in the UK and one of the first in the world to pilot the technology as part of its cervical screening programme. Health experts said the new technology could be instrumental in ensuring earlier detection of pre-cancerous cells and cancer cells and has the potential to save lives. The pilot is using a digital cytology system, the GeniusTM Digital Diagnostics System, from women's health company Hologic. For the pilot programme the system will create digital images of cervical smear slides from samples that have tested positive for Human Papilloma Virus (HPV).

There is little argument about whether or not machine learning models are useful for applying to social systems. But if we take seriously George Box's dictum, or indeed the even older one that "the map is not the territory' (Korzybski, 1933), then there has been comparatively less systematic attention paid within the field to how machine learning models are wrong (Selbst et al., 2019) and seeing possible harms in that light. By "wrong" I do not mean in terms of making misclassifications, or even fitting over the'wrong' class of functions, but more fundamental mathematical/statistical assumptions, philosophical (in the sense used by Abbott, 1988) commitments about how we represent the world, and sociological processes of how models interact with target phenomena. This paper takes a particular model of machine learning research or application: one that its creators and deployers think provides a reliable way of interacting with the social world (whether that is through understanding, or in making predictions) without any intent to cause harm (McQuillan, 2018) and, in fact, a desire to not cause harm and instead improve the world, 1 for example as most explicitly in the various "{Data [Science], Machine Learning, Artificial Intelligence} for [Social] Good" initiatives, and more widely in framings around "fairness" or "ethics." I focus on the almost entirely statistical modern version of machine learning, rather than eclipsed older visions (see section 3). While many of the limitations I discuss apply to the use of machine learning in any domain, I focus on applications to the social world in order to explore the domain where limitations are strongest and stickiest.