Perplexity Research with Harvard Shows AI Agents Cut Task Time and Cost

PerplexityPerplexity

Perplexity, in collaboration with Harvard Business School, published new research on its Computer autonomous agent. The study found that workers using Computer completed tasks in 87% less time and at 94% lower cost than using Search alone, demonstrating how agents expand the scope and efficiency of knowledge work.

Perplexity, in collaboration with Harvard Business School, published new research on its Computer autonomous agent, detailing its impact on knowledge work. The study found that workers using Computer completed tasks in 87% less time and at 94% lower cost compared to using Perplexity Search alone, with higher user satisfaction.
Task completion time reduction
87%
Task cost reduction
94%
Machine execution time increase
48x
New tasks unlocked
23% of Computer queries
Next-turn dissatisfaction reduction
55%
Top task category
Research and Analysis (25.8%)

The research highlights a shift to agentic systems, with Computer performing 48 times more machine execution per session than Search. It also makes 12 times more external tool calls per session, indicating greater reach and action capability. This autonomy allows users to delegate complex workflows and expand task scope.

Perplexity Computer shifts the user's role from manual execution to supervision, enabling goal specification and output review while the agent handles intermediate steps. This expands the task frontier, allowing users to take on broader, deeper work, including tasks requiring expertise across multiple domains.

Perplexity data highlighting AI agent impact on knowledge work efficiency, task completion time, and operational cost savings.
View the full update on research.perplexity.ai
Perplexity
Perplexity
@perplexity_ai
X

We published new research with Harvard on the shift from chat interfaces to autonomous agents like Computer. Over 3 months, findings show workers using Computer finish tasks in 87% less time at 94% lower cost than Search alone, with higher satisfaction. https://t.co/qmcUqcj8CI https://t.co/R4oTLavC6T

71retweets495likes
View on X

Still wondering? A few quick answers below.

Perplexity Computer is a general-purpose autonomous agent orchestrator that autonomously plans, reasons, uses tools, and takes multi-step actions to achieve user-specified goals across complex environments and time horizons. It shifts AI usage from conversational responses to autonomous task execution.

The research found that workers using Perplexity Computer completed tasks in 87% less time and at 94% lower cost compared to using Perplexity Search alone. These efficiency gains were observed across various knowledge work domains.

Perplexity Computer expands the scope of work, enabling users to take on tasks that are both broader and deeper. It allows for work that crosses occupational boundaries and requires expertise across multiple knowledge domains, with 23% of Computer queries engaging tasks never attempted with Search.

Perplexity Computer performs significantly more autonomous machine work, averaging 26 minutes of machine execution per session compared to 33 seconds for Perplexity Search. It also makes 12 times more external tool calls per session, indicating greater reach and action capability.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Share this update