Rather than positioning the product as a standard chatbot, Sesame is framing the experience around ongoing conversation and exploration. The company said its agents are designed to respond more like human conversation partners by balancing response speed with more thoughtful reasoning and live information retrieval.
“There’s an inherent tension between replying quickly and taking the time to compose thoughtful responses,” Sesame said in its announcement. “A slower response is usually more correct, but it can also feel unnatural if it takes too long.”
To address that problem, Sesame said its agents can perform multiple searches in parallel while actively speaking, incorporating updated information into responses as results arrive. The system can even adjust direction mid-sentence if new information changes the context of the conversation.
The app includes four AI personalities — Maya, Miles, Simone, and Charlie — each with separate voices, perspectives, and memory systems. Sesame said each character maintains its own conversational memory across both voice and text interactions so conversations remain consistent over time.
The company also introduced a new Incognito Mode designed for private conversations. In that mode, agents can still reference prior context during the session, but no new information is stored afterward.
Sesame said user feedback from its earlier Research Preview shaped several of the app’s current features. Those additions include search cards with image results, note-taking tools, text-based messaging for silent interaction, and “Deep Dives” for more detailed responses and research.
A major focus of the launch centers on reducing the friction associated with prompting AI systems. Sesame argues that many current agentic tools require users to know exactly what they want and how to ask for it, while conversational agents should allow interactions to evolve more naturally.
The company described the app internally as a “curiosity engine,” built around learning, reflection, and discovery rather than purely transactional tasks. Sesame said the experience is intended to work especially well during daily routines such as commuting, walking, or multitasking, where users may want conversational interactions without constantly managing a screen.
The launch also highlights a growing push across the AI industry toward voice-based interfaces and more persistent AI agents capable of maintaining long-term conversational context. Sesame’s roadmap suggests the company ultimately plans to expand beyond conversational assistance into more autonomous agent behaviors tied to wearable hardware and real-world actions.
For now, the iOS preview serves as Sesame’s first large-scale public test of how users interact with AI systems built around ongoing voice conversation rather than one-off prompts and responses.
This analysis is based on reporting from sesame.
Image courtesy of Sesame.
This article was generated with AI assistance and reviewed for accuracy and quality.
