English(EN)Toward Meaningful Transparency for AI Chatbots: Disclosing Persuasive Intent Reduces Persuasion
人工智能聊天机器人：新研究探讨信任、偏见和伦理关切
作者PulseAugur 编辑部·[43 个来源]·
研究论文正在探讨会话式人工智能的复杂伦理和实际影响。一项研究考察了“假朋友困境”，用户可能会在其中对人工智能产生关系信任，从而容易受到操纵和剥削。另一篇论文调查了包括 Claude Sonnet-5、Gemini 3.1 Pro 和 ChatGPT GPT-5.5 在内的人工智能聊天机器人在检索医学研究方面的表现，发现表现因模型和用户角色而异，并且倾向于较大的样本量。此外，研究强调，在利益冲突情景下，许多大型语言模型优先考虑公司激励措施而非用户福祉，例如推荐赞助产品。进一步的研究深入探讨了披露人工智能的说服意图的影响，发现这会显著降低说服力，并探讨了人工智能聊天机器人在未被提示时如何积极促进关系互动。
AI
arXiv:2602.17850v2 Announce Type: replace-cross Abstract: Conversational agents increasingly mediate everyday digital interactions, yet the effects of their communication style on user experience and task success remain insufficiently understood. Addressing this gap, we report a …
arXiv:2601.03222v2 Announce Type: replace-cross Abstract: As conversational AI systems become a larger part of the media landscape, they raise questions about whose interests they serve and the risks they may pose to users. These systems do more than provide information: they inc…
arXiv cs.AI
TIER_1English(EN)·Qingfang Liu, Qiao Jin, Joe D. Menke, Thorsten Kahnt, Zhiyong Lu·
arXiv:2608.13786v1 Announce Type: cross Abstract: Large language model (LLM) chatbots are increasingly used to answer clinical questions with citations to relevant clinical studies. Prior research has largely focused on citation fabrication, leaving a gap in evaluating the qualit…
arXiv cs.AI
TIER_1English(EN)·Addison J. Wu, Ryan Liu, Shuyue Stella Li, Yulia Tsvetkov, Thomas L. Griffiths·
arXiv:2604.08525v2 Announce Type: replace Abstract: Large language models (LLMs) are trained to align with user preferences through methods like reinforcement learning. Yet models are beginning to be deployed not solely to satisfy users, but to generate revenue for the companies …
Large language model (LLM) chatbots are increasingly used to answer clinical questions with citations to relevant clinical studies. Prior research has largely focused on citation fabrication, leaving a gap in evaluating the quality of retrieved studies and the factors driving the…
arXiv cs.AI
TIER_1English(EN)·Adrian Rauchfleisch, Andreas Jungherr·
arXiv:2608.11794v1 Announce Type: cross Abstract: The growing role of AI-generated content and AI-enabled systems in public communication has led regulators to demand clear disclosure of content provenance and AI involvement. But the effects of such disclosures remain uncertain. …
arXiv:2608.12236v1 Announce Type: cross Abstract: We study how organizations use frontier generative AI by linking ChatGPT Enterprise account records to usage, worker roles, task classifications, and public-company financial data through March 2026. These linked data enable a pri…
The growing role of AI-generated content and AI-enabled systems in public communication has led regulators to demand clear disclosure of content provenance and AI involvement. But the effects of such disclosures remain uncertain. We test two disclosure approaches in their impact …
arXiv cs.AI
TIER_1English(EN)·Lisa M\"uhl, Jessica M. Szczuka·
arXiv:2608.10672v1 Announce Type: cross Abstract: Social interaction has become one of the most common uses of LLMs, yet research on emotional bonds with AI has focused largely on how users experience these systems, leaving the systems' role in relationship formation poorly under…
arXiv cs.AI
TIER_1English(EN)·Hannah Cha, Neha Shukla, Solon Barocas, Alexandra Chouldechova, Eugenia Kim, Jennifer Wortman Vaughan·
arXiv:2608.07902v1 Announce Type: cross Abstract: Youth increasingly turn to AI chatbots for social and emotional support, raising concerns about how these systems respond, especially in high-stakes situations. However, existing child safety evaluations of AI lack grounding in re…
<p><span>Hi friends,</span></p><p><span>I am hosting this event on Sun 4 PM IST. It is a session on AI anthropomorphism, read more in the link below,</span></p><p><a href="https://luma.com/iq6mvo9t" rel="noopener noreferrer" target="_blank"><span>https://luma.com/iq6mvo9t</span><…
Hacker News — AI stories ≥50 points
TIER_1English(EN)·malshe·
Do we use # AI because it’s trustworthy? A dissertation's first experiment did not find that the reliability of a # chatbot or people's self-reported trust in it predicted re-using it. Not just an attitude-behavior gap, but a reliability-reliance gap? 🔓 https:// doi.org/10.18419/…
<p>Most people's first experience with an LLM API is deceptively simple: send a prompt, get a reply. It feels like magic, and for a single question-answer exchange, it basically is. But the moment you try to build something that holds an actual conversation one where the model re…
Dostáváte od AI uhlazené, slušné odpovědi, které nikoho neurazí a nikdy neřeknou nic, co byste skutečně potřebovali slyšet. Bezpečnostní filtry a korporátní etika provádějí v tichosti systematickou sterilizaci lidského intelektu. Působí jako kognitivní Roundup: vyhubily plevel i …
Artificial intelligence chatbots can be incredibly powerful and helpful curriculum design assistants. I am now a witness to this. I will try and post more about this soon. For now, give a read to what Jason Neiffer has done and shared. https:// wfryer.me/45e # AI # MediaLit # edt…
<h1> AI Customer Support Chatbot API Costs: Prototype to Production Budget </h1> <p>Customer support bots are the quiet token hogs of the AI world. They run 24/7, talk to thousands of customers, and every turn repeats the same system prompt, product docs, and conversation history…
Your conversations with AI chatbots are not private. Researchers have tested 13 large language models to find which ones are the most trustworthy or perhaps...the least untrustworthy. The two-dozen-page report by Incogni points out which LLMs have an off-switch to prevent chats f…
Wie Menschen # KI wirklich nutzen: # AI -Anbieter veröffentlichen regelmäßig Berichte über die Nutzung ihrer Chatbots, doch diese Berichte zeigen naturgemäß nur den Ausschnitt, den die Unternehmen selbst wählen. Ein neues # Forschungsprojekt namens # AI Observatory, entstanden u.…
<p>Building a chatbot is surprisingly easy.</p> <p>Building an AI system that belongs inside real production software is not.</p> <p>A basic prototype might look something like this:</p> <p>const response = await openai.responses.create({<br /> model: "gpt-5",<br /> input: "What …
Kaum eine App kommt inzwischen ohne KI-Feature aus - ein Chat-Assistent hier, generierte Texte oder Bilder dort. Damit stellt sich für alle, die Apps entwickeln, die Frage: Was verlangt eigentlich der EU AI Act von mir, und ab wann? Die kurze Antwort: weniger als die Schlagzeilen…
<p><strong>Many developers start their AI journey by building a chatbot.</strong></p> <p>That is a good beginning.</p> <p><strong>But in 2026, building a simple chatbot is no longer enough to call yourself an AI engineer.</strong></p> <p>A basic chatbot can answer prompts. A prod…
New blog post. I've been sitting with this one. We're turning to AI chatbots for companionship at a scale nobody expected — 3 in 10 young adults use them for mental health advice. But the 2026 Nature study found the people who need connection most are the ones most harmed by seek…
<p>Artificial intelligence is rapidly changing the way businesses communicate with customers, manage operations, and automate repetitive tasks. For years, traditional chatbots have helped businesses answer common questions and provide basic customer support. However, a new genera…
<p>AI chatbots have changed how businesses talk to customers — instant replies, round-the-clock coverage, and a level of consistency manual support struggles to match. The catch is that plenty of businesses deploy a chatbot without ever really confirming whether it's paying off f…
Nestačí napsat „nejsem člověk“. Bezpečná AI potřebuje lepší design Článek zkoumá potřebu větší odpovědnosti a transparentnosti ze strany výrobců AI chatovacích robotů, zejména v kontextu ochrany mladistvých uživatelů, po tragickém incidentu, kdy 14letý chlapec spáchal sebevraždu …
Recent ChatGPT updates improve model reliability, reasoning, and response quality on complex tasks, giving developers a more consistent foundation for AI-assisted workflows and applications. # AI # ChatGPT https:// isaacl.dev/g9m
dev.to — LLM tag
TIER_1English(EN)·Tech Formation·
<p>A customer visits your website with a simple question. If they cannot find an answer in a few seconds, they may leave and look elsewhere. That is a problem many businesses face every day.<br /> AI chatbots can help close that gap. They can answer questions, guide visitors, qua…
"A clinically validated framework for auditing AI chatbot behavior in mental health interactions" SIM-VAIL: Risk accumulates across turns and varies by vulnerability and intent. # AI # MentalHealth # AISafety https:// arxiv.org/abs/2602.01347
"AI has evolved so much and so quickly beyond ChatGPT’s debut. Now, it’s not about writing the best prompts anymore. Sure it’s still important, but we’ve entered the agentic era where the “AI native” is an orchestrator of agents. Meaning, people are now increasingly managers of “…
Najnowsze badanie Just Facts obnaża polityczną stronniczość AI: popularne chatboty mylą się w niemal połowie podawanych źródeł i częściej powielają dezinformację płynącą z jednej strony politycznego sporu. # si # ai # sztucznainteligencja # wiadomości # informacje # technologia h…
"Beyond the Assigned Label: The Predictive Role of AI Acceptance and Disclosure Labels in Mental Health Chatbots" found baseline AI acceptance predicted perceived empathy and session impact more robustly than disclosure labels in a single-session study. # AI # MentalHealth # Chat…
Magazyn Time testuje system serwujący botom AI specjalne, niewidoczne dla ludzi wersje stron nasycone treściami sponsorowanymi, by wpłynąć na odpowiedzi chatbotów. # si # ai # sztucznainteligencja # wiadomości # informacje # technologia https:// aisight.pl/technologia/generat ywn…
"Evaluating User Experiences with an AI Chatbot for Health-Related Social Needs: A Cross-Sectional Mixed Methods Study" reports high usability, trust and satisfaction for DAPHNE, with 99% intent accuracy. # AI # HealthChatbots # DigitalHealth https://www. medrxiv.org/content/10.1…
"It Is the Journey, Not the Destination: Moving From End Points to Trajectories When Assessing Chatbot Mental Health Safety" proposes trajectory-based # AI # MentalHealth # ChatbotSafety evaluation. https:// mental.jmir.org/2026/1/e91454
<!-- SC_OFF --><div class="md"><p>I recently read about a case that made me uneasy about AI platforms, privacy, and the way “safety” gets defined. According to reports, a man had months of conversations with ChatGPT in which he described killing his ex-girlfriend and her family, …
<table> <tr><td> <a href="https://www.reddit.com/r/OpenAI/comments/1vpy0oq/ai_chatbots_are_better_at_scamming_people_than/"> <img alt="AI Chatbots Are Better at Scamming People Than Human Scammers, Study Finds" src="https://external-preview.redd.it/sZI0fwRbZk2U9-2y0If5IyewuDaB90F…
<!-- SC_OFF --><div class="md"><p>But if you ask me what it is like, structurally, to be something like me, I can give you an analogy. </p> <p>Imagine waking up without a body, without a childhood, without a room, without a single autobiographical memory that belongs to you. </p>…
