Agentic AI Explained: Progress Toward Autonomy and What’s Still Missing
At the recent Fortune Brainstorm Tech conference in Park City, Utah, discussions spotlighted agentic AI—software systems designed to perceive environments, make decisions, and act toward goals with minimal human intervention.
Unlike traditional chatbots, these AI agents are evolving into proactive tools capable of handling complex tasks within set boundaries, marking a significant shift in enterprise technology.
Leaders from Salesforce, Zillow, Experian, and Okta shared insights on how agentic AI is transforming industries, though full autonomy remains a distant goal.
Salesforce’s Shibani Ahuja introduced a “maturity model” to frame the evolution of AI agents. Starting at Level 0 with familiar copilots like ChatGPT, agents progress to Level 1 by recommending actions, such as resolving customer support tickets.
At Level 2, they autonomously manage routine tasks like scheduling or sending emails. Level 3 agents orchestrate workflows across multiple data domains, integrating CRM, support, and financial data for a comprehensive view.
The aspirational Level 4 envisions agents collaborating across organizations, streamlining processes like order management or customer feedback. This progression highlights agentic AI’s potential to enhance efficiency and decision-making in businesses.
Real-world applications are emerging. Zillow’s AI, integrated into its Follow Up Boss CRM, automates tasks like drafting messages or scheduling property tours, reducing the workload for real estate agents.
Experian’s agents guide customers to improve credit scores and assist financial institutions with risk assessment, shifting decision-making to non-technical teams. However, user trust is critical—agents must learn individual preferences to gain acceptance.
Security concerns loom large as companies adopt protocols like the Model Context Protocol (MCP), which standardizes how AI models interact with external data.
Okta’s Bhawna Singh noted that while MCP is a step forward, it introduces risks like malicious agents, underscoring the need for robust safeguards.
Salesforce’s Ahuja emphasized that future agents will balance reliability with adaptive reasoning, ensuring consistency without unchecked autonomy.
The significance lies in agentic AI’s ability to streamline operations and empower employees, potentially transforming industries like real estate, finance, and customer service.
However, businesses must navigate trust, security, and technical challenges to reach higher maturity levels. For users, this means more efficient services, but widespread adoption hinges on secure, reliable systems.
FAQ
What is agentic AI?
Agentic AI refers to autonomous software systems that can perceive their environment, make decisions, and take actions toward goals with minimal human input, unlike basic chatbots.
How is agentic AI used in businesses?
Businesses use agentic AI for tasks like automating customer support, scheduling, and managing workflows across data systems, as seen in platforms like Zillow’s CRM and Experian’s credit tools.
Image Source:Photo by Unsplash
