Sierra’s Bret Taylor says the era of clicking buttons is over
Bret Taylor, co-founder and CEO of Sierra, a startup focused on building AI agents for enterprise customer service, predicts a significant shift in how users interact with software. He believes that the traditional, click-based interface is becoming obsolete, replaced by natural language interactions.
Ghostwriter: The Agent Builder
Sierra recently launched Ghostwriter, an AI agent designed to create other AI agents. This "agent as a service" tool aims to eliminate the need for users to navigate complex software interfaces. Instead, users can simply describe their needs in natural language, and Ghostwriter will autonomously build and deploy a specialized agent to fulfill the request.
The Problem with Traditional Software
Taylor argues that many enterprise tools are underutilized because they are complex and only needed for specific, infrequent tasks. He uses Workday as an example, noting that employees typically only interact with it during onboarding or open enrollment. The future, he suggests, lies in using natural language to accomplish tasks without ever needing to learn or use a software interface.
Rapid Deployment and Business Value
Sierra is leveraging Ghostwriter to achieve rapid agent deployment. Taylor cited an example where an agent for Nordstrom was implemented in just four weeks. He emphasizes that companies are not looking to build software, but rather to find solutions to their problems, a goal that AI agents can directly address.
Market Traction and Valuation
Sierra has demonstrated significant market traction, reaching $100 million in annual revenue run rate (ARR) in less than 21 months. The company was last valued at $10 billion after raising a $350 million round.
The Reality of AI Agents
While Taylor's vision is compelling, the article notes a counterpoint: many companies, including Sierra and legal AI startup Harvey, still rely on "forward-deployed" engineers. These engineers are crucial for continuously updating and fine-tuning AI agents to ensure they function as intended. This suggests that fully autonomous, interface-free interactions are still a work in progress.
Key Takeaways:
- Shift from Interfaces to Language: The future of software interaction will likely move away from clicking buttons and towards natural language prompts.
- AI Agents as Solutions: AI agents can directly address business needs by providing solutions rather than requiring users to learn complex software.
- Accelerated Development: Tools like Sierra's Ghostwriter can significantly speed up the creation and deployment of specialized AI agents.
- Current Limitations: The widespread adoption of fully autonomous AI agents is still challenged by the need for human oversight and continuous fine-tuning.
- Focus on Value: The ultimate goal for businesses is problem-solving, and AI agents are positioned to deliver this value more directly.