This study explored how “a range of current AI systems have learned how to deceive humans”. Extracts: · “One part of the problem is inaccurate AI systems, such as chatbots whose confabulations are often assumed to be truthful by unsuspecting users” · “It is difficult to talk about deception in AI systems without psychologizing them. In humans,… Continue reading AI deception: A survey of examples, risks, and potential solutions
Tag: llm
Agentic Misalignment: How LLMs could be insider threats (Anthropic research)
AI and malicious compliance. This research from Anthropic has done the rounds, but quite interesting. In controlled experiments (not real-world applications), they found that AI models could resort to “malicious insider behaviors when that was the only way to avoid replacement or achieve their goals—including blackmailing officials and leaking sensitive information to competitors”. Some extracts:… Continue reading Agentic Misalignment: How LLMs could be insider threats (Anthropic research)
Large Language Models in Lung Cancer: Systematic Review
This systematic review of 28 studies explored the application of LLMs for lung cancer care and management. Probably few surprises here. And it’s focused mostly on LLMs, rather than specialised AI models. Extracts: · The review identified 7 primary application domains of LLMs in LC: auxiliary diagnosis, information extraction, question answering, scientific research, medical education, nursing… Continue reading Large Language Models in Lung Cancer: Systematic Review
From transcript to insights: Summarizing safety culture interviews with LLMs
From transcript to insights: summarizing safety culture interviews with LLMs How well does OpenAI o1 work for summarising ‘safety culture’ interviews, and how does it compare to human notes? This study did just that. Extracts: · They assessed correctness via exhaustiveness (comparison of LLM claims vs human interviewer notes), consistency (comparison of LLM claims between subsequent… Continue reading From transcript to insights: Summarizing safety culture interviews with LLMs
Practice With Less AI Makes Perfect: Partially Automated AI During Training Leads to Better Worker Motivation, Engagement, and Skill Acquisition
How does AI use in training improve, or impact, skill acquisition? This study manipulated training protocols with varying levels of AI decision-making automation, among 102 participants during a quality control task. Extracts: · “Partial automation led to the most positive outcomes” · “Participants who were trained with the fully automated version of the AIEDS had a significantly… Continue reading Practice With Less AI Makes Perfect: Partially Automated AI During Training Leads to Better Worker Motivation, Engagement, and Skill Acquisition
Safe As 33: Is ChatGPT bullsh** you? How Large Language models aim to be convincing rather than truthful
Large Language Models, like ChatGPT have amazing capabilities. But are their responses, aiming to be convincing human text, more indicative of BS? That is, responses that are indifferent to the truth? If they are, what are the practical implications? Today’s paper is: Hicks, M. T., Humphries, J., & Slater, J. (2024). ChatGPT is bullshit. Ethics and… Continue reading Safe As 33: Is ChatGPT bullsh** you? How Large Language models aim to be convincing rather than truthful
Can chatbots provide more social connection than humans?
Can chatbots provide more social connection than humans? Possibly, providing that they don’t “claim too much humanity”. Three study protocols with 801, 201 and 401 had participants engage with AI social chatbots. They note that the long-term consequences of social chatbot use is unknown, but is important to study since “hundreds of millions of people… Continue reading Can chatbots provide more social connection than humans?
Enhancing AI-Assisted Group Decision Making through LLM-Powered Devil’s Advocate
Can LLM’s effectively play as devil’s advocate, enhancing group decisions? Something I’ve been working on lately is AI as a co-agent for cognitive diversity / requisite imagination. Here’s a study which explored an LLM as a devil’s advocate, and I’ll post another study next week on AI and red teaming. [Though this study relied on… Continue reading Enhancing AI-Assisted Group Decision Making through LLM-Powered Devil’s Advocate
Endoscopist De-Skilling after Exposure to Artificial Intelligence in Colonoscopy: A Multicenter Observational Study
Does AI use contribute to de-skilling? Probably, according to this study of endoscopists. This study compared >1.4k patient outcomes who underwent non-AI assisted colonoscopy before and after AI implementation. Background: · A recent meta-analysis of 20 randomised trials “showed an absolute 8.1 % increase in ADR [Adenoma detection rate] with the use of AI during colonoscopy.5… Continue reading Endoscopist De-Skilling after Exposure to Artificial Intelligence in Colonoscopy: A Multicenter Observational Study
The impact of generative AI on critical thinking skills: a systematic review, conceptual framework and future research directions
The impact of generative AI on critical thinking skills: a systematic review, conceptual framework and future research directions How do generative AI (GenAI) models affect critical thinking skills? This systematic review unpacked 68 studies to explore the good and the bad. GenAI are “machine-learning algorithms, usually transformer-based large-language models (LLMs), that generate new text, code… Continue reading The impact of generative AI on critical thinking skills: a systematic review, conceptual framework and future research directions
SafetyInsights.org 