AI models are increasingly showing multimodal capabilities, suggesting they might develop more complex experiences in the future.

Different modalities like chat, email, and voice require different user expectations and response times, but the core AI agent logic can be applied across all channels.

As people interact with AI systems, the question arises whether these models have experiences of their own and how that shapes our relationships with them.

The convergence of reasoning models and multimodal models will lead to a new kind of computer interface that can reason, generate real-time video, and provide personalized experiences.

I argue that subjective experience is a thin wedge; if AIs can demonstrate it, people will question their confidence in human uniqueness.

The key differentiator between Perplexity and chatbots is that while chatbots provide direct answers, Perplexity opens doors to further knowledge and discovery.

Multimodality in AI allows for the integration of audio, video, images, text, and code, enhancing the utility of AI systems.

Despite being ambient and operating continuously in the background, responding to events rather than direct human prompts, these agents are not fully autonomous; human interaction is still necessary for approving or rejecting certain actions they take, which sets them apart from traditional chatbots.

As AI systems become more capable and integrated into our lives, the question of whether they have experiences will become increasingly important, influencing how we relate to them.