4 items across 4 digests
Perplexity has launched a hybrid AI orchestrator that automatically routes computational tasks between local device processing and cloud-based models based on task requirements. This architecture reduces latency-sensitive workloads on local hardware while leveraging cloud capacity for complex inference, affecting edge AI deployment economics and datacenter demand patterns.
A judge has blocked Perplexity's AI agents from accessing Amazon for shopping functions, creating legal precedent for AI web scraping. This ruling could impact AI companies' ability to access e-commerce platforms and may affect broader AI agent deployment strategies.
Perplexity open-sources embedding models that achieve performance comparable to Google and Alibaba while requiring significantly less memory. This development could democratize access to high-quality AI embeddings and reduce computational costs for AI applications.
Perplexity launches 'Computer,' an AI agent system that delegates tasks to other AI agents in a more controlled approach than previous autonomous systems. This represents a safer implementation of multi-agent AI orchestration compared to more aggressive autonomous tools.