Adept

Useful General Intelligence — Danielle Perszyk, Amazon AGI

Aug 2, 2025 · 19:58

Danielle Perszyk, a cognitive scientist at Amazon's AGI SF Lab, argues that general-purpose AI agents like Nova Act must be built to augment human intelligence rather than replace it, drawing parallels to how human cognition relies on controlled hallucinations and social co-evolution. She explains that current agents struggle with basic computer tasks like clicking, typing, and scrolling, so Nova Act combines a specialized version of Amazon Nova trained via RL for UI reliability with an SDK that allows developers to string together atomic actions (act calls) into complex workflows. The lab's vision centers on grounding agents in a shared environment and eventually giving them models of human minds, which requires a common language for humans and agents to reciprocally level up intelligence. Demos with teammates Carolyn (searching apartments) and Fjord (extracting data with Python and parallel browsers) show how Nova Act enables three-line code starts and granular control, positioning the product as a primitive for a cognitive technology that aligns representations and stabilizes human thinking.

The Web Browser Is All You Need - Paul Klein IV, Browserbase

Jun 20, 2025 · 17:31

Paul Klein IV, founder of Browserbase, argues that the web browser is the default MCP server for the rest of the internet, serving as the essential bridge between AI agents and legacy websites that lack APIs. He distinguishes between web agents (one prompt to many actions, e.g., OpenAI's Operator) and browser tools (one action per prompt), noting that both rely on vision-driven or text-based approaches to parse pages. Klein explains that Browserbase offers a horizontal MCP server for browsing, enabling automation of any website, and stresses the need for custom evals and observability to track agent behavior. In a live demo, he shows an agent navigating sfpca.org to find a dog for adoption, handling unexpected modals. He also addresses captchas, advising good citizenship and hinting at future agent authentication solutions.

Climbing the Ladder of Abstraction: Amelia Wattenberger

Nov 3, 2023 · 16:47

Amelia Wattenberger, a designer at Adept, argues that AI can augment knowledge work by enabling users to 'zoom out' on information across different levels of abstraction, much like Google Maps hides detail at higher zoom levels. She reframes augmentation as composed of smaller automations, using the example of spreadsheets automating calculations so accountants can focus on analyzing numbers. Wattenberger demonstrates a prototype that uses LLMs to summarize paragraphs of Peter Pan into single sentences at higher zoom levels, and sketches how Adept's AI could extract key factors like Wi-Fi speed and walking distance from Airbnb and hotel listings, presenting them in customizable views and even a scatter plot for comparing 50 listings at once. The talk emphasizes that structured interfaces around AI models—not just chatbots—will transform how we work, and that Adept is building AI to automate tedious tasks while keeping users in control.