Tag
Omio is using OpenAI's ChatGPT and Codex to build conversational travel booking experiences and transform internal operations, moving toward an AI-native approach.
A discussion on whether people find it easier to discuss personal topics with AI or humans, noting that AI offers a non-judgmental, always-available ear but lacks genuine human experience.
A user reports that gpt-5.5 via the codex agentRuntime silently drops tool calls on short conversational turns, resulting in no response despite the model generating text. The issue is specific to gpt-5.5 and is fixed by switching to deepseek/deepseek-v4-pro, suggesting a regression.
The article challenges the conventional view of support chatbots as stateless FAQs and explores how rethinking their design can lead to more effective, stateful conversational agents.
Pinterest launches an experimental AI shopping app called 'Ask Pinterest' that uses conversational AI and its Taste Graph for personalized product recommendations, alongside new AI ad tools for marketers.
OpenAI plans to release GPT-Bidi-1, its next-generation voice model that can listen and speak simultaneously, handle interruptions, and enable more natural conversations.
This paper introduces Turing-RL, a reinforcement learning approach that uses Turing test-based rewards to train language models to generate responses indistinguishable from human users in conversational and forum settings, outperforming baseline methods.
T-Mem is a new long-term conversational memory architecture that enables both descriptive and associative recall, covering scenarios where query and memory share surface features and those where they are connected by latent semantic arcs. It reaches state-of-the-art on the LoCoMo and LoCoMo-Plus benchmarks.
The author argues that current AI excels at processing transcript language but misses non-verbal cues like hesitation and tone, highlighting a gap between understanding language and understanding human communication.
D-ID introduces agentic videos, interactive AI-generated videos that can respond to users.
DoorDash announces 'Ask DoorDash,' an AI chatbot that lets users order food and groceries using text prompts or photos, rolling out on iOS in select regions.
The paper presents a pipeline that maps student questions from a conversational AI teaching assistant to curriculum topics using a few-shot text classifier and a GPT-4-extracted prerequisite knowledge graph, achieving 80% accuracy on 1,340 question events and correlating with self-reported difficulty.
Apple announced 'Siri AI' at WWDC, a more conversational voice assistant with deeper integration across apps and personal context, rolling out this fall alongside Google-powered AI model updates.
Apple unveiled a major overhaul of Siri at WWDC 2026, transforming it from a voice assistant into a conversational AI chatbot called Siri AI, with capabilities like web grounding, device context, and a dedicated app.
AI voice agents like LuMay Voice Agent and Voxentis.ai are accelerating the replacement of traditional call centers due to advances in conversational AI, scalability, and performance consistency, driving hybrid AI-human models across industries such as healthcare, real estate, and e-commerce.
This paper presents COVA-X, an expanded synthetic multi-turn conversation dataset for smishing detection, and shows that Longformer now outperforms XGBoost, confirming that transformer models benefit from larger training corpora.
τ-Rec is a verifiable benchmark for agentic recommender systems that replaces subjective LLM-as-a-judge evaluations with verifiable rewards and controlled dialogue constraints, revealing steep reliability cliffs across leading models where even the best achieves only ~57% pass@1.
Researchers introduce BenSyc, the first benchmark for evaluating conversational sycophancy in Bengali social contexts, finding that LLMs struggle to distinguish empathetic support from validation and escalation, achieving only ~61% Macro-F1.
This paper proposes a training-free, CPU-only retrieval method that fuses BM25 lexical scores with late-interaction dense scores for conversational memory retrieval, achieving up to +17.2 points improvement on LoCoMo Hit@1 over late interaction alone across six encoders. The study provides controlled ablations on pooling operators, reranker effects, and benchmark robustness, framing the gain as a division of labor between dense and lexical signals.
Researchers from University of Toronto and Vector Institute propose Segment Tree Memory (SegTreeMem), a memory architecture for long-horizon conversational agents that preserves temporal order using a hierarchical segment tree structure for both online construction and retrieval. Experiments across three datasets show nearly 20% improvement in LLM-judge accuracy over non-temporal tree baselines.