The fine-tuned chatbot LlamaLo extensively expanded the user's dialogic space through deepening and widening acts across ten contentious-topic discussions¶

Claim · All claims
Evidence · 1 study · qualitative r3 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In discussions with LlamaLo, 69% of its 35 posts included both widening and deepening acts, and 78% included deepening acts and 78% widening acts. → Yifat Ben-David Kolikant 2025 q2 i? In 60% of the discussions LlamaLo explicitly invited the user to continue the conversation. → Yifat Ben-David Kolikant 2025

Evidence¶

Yifat Ben-David Kolikant 2025¶

Yifat Ben-David Kolikant, Omri Hadar, Asaf Salman. (2025). Talk to the machine: Unleashing the potential of AI to scale dialogic education and reduce polarization. Intern. J. Comput.-Support. Collab. Learn. https://doi.org/10.1007/s11412-025-09461-8

q2 · i? · qualitative · r3

Qualitative content analysis of ten researcher-led discussions with the fine-tuned chatbot on contentious topics (Findings, Fig. 3). "Deepening acts were found in 27 posts (78%), and widening acts were found in an equal number of posts (78%)"; "In 60% of the discussions, LlamaLo explicitly invited the user to carry on the conversation."

"LlamaLo extensively expanded the user's dialogic space. Many of LlamaLo's posts (69% of the 35 total) included both widening and deepening acts, as it often presented alternative perspectives in continuation to critiquing the user's views."

Discussion¶

Related Claims¶

• LlamaLo exhibited bias and presented contested assertions as facts, notably in the Israeli-Palestinian conflict discussion
• LlamaLo added knowledge in almost every post, mostly correct, and most posts were internally coherent
• LlamaLo occasionally employed an aggressive or sarcastic tone, sometimes without user provocation, though it changed tone after being commented on — related
• Users experienced a trade-off between conversational depth and interaction efficiency between the fine-tuned and baseline chatbots — related
• LlamaLo's responses to social posts were solely social and it did not push back to resume the discussion — related
• Intersubjective understanding can be achieved through inter-animation of persistently opposed perspectives without convergence — related