Salt Security’s latest report, Securing the Future of Agentic AI: Building Consumer Trust through Robust API Security, warns that without effective API discovery, governance, and protection, the very tools designed to enhance customer engagement could instead expose organizations to cyberattacks or data breaches. The findings also point to a widening trust gap, as consumers grow hesitant to share personal data with businesses using agentic AI for external interactions due to security fears.
Because APIs power AI agents, with the ability to make requests, retrieve data and interact across different platforms, the common thread towards improving confidence in agentic AI interactions is API security. The Securing the Future of Agentic AI report proposes that once security is strengthened, consumer trust will follow and allow agentic AI to reach its full business potential.
The unique report delved into both sides of the agentic AI equation – those organizations already using it and consumers that are encountering it.
Key Highlights from Businesses Using Agentic AI:
- Over half (53%) of organisations using agentic AI say they are already deploying it, or plan to, for customer-facing roles
- Nearly half (48%) of organizations currently use between 6 and 20 types of AI agents and 19% deploy between 21 and 50; 37% of organizations report that 1–100 AI agents are currently active within their systems; and almost a fifth (18%) host between 501–1000 agents
- Only 32% conduct daily API risk assessments
- 37% have a dedicated API security solution
- Just 37% have a data privacy team overseeing AI initiatives
- 7% assess API risk monthly or less
Key Highlights from Consumers:
- 64% of consumers have interacted with AIÂ chatbots more frequently in the past year
- 80% of those consumers have shared personal information during these interactions
- 44% say they’ve felt pressured to share information just to complete a task
- Only 22% of consumers are comfortable sharing data with AI agents, compared to 37% who trust interactions over the phone and 54% in person
- 62% believe AI agents are easier to trick than humans
AI Agent Use is Exploding, But Consumers Aren’t Convinced
The report revealed that over half of organisations already using agentic AI say they deploy it or plan to for customer-facing tasks. Meanwhile, 64% of consumers report encountering AI chatbots more frequently than a year ago, and four in five who interact with them have shared personal details in the process.
Yet, trust remains a significant hurdle. Just 22% of consumers feel comfortable sharing data with chatbots, compared to 54% in person and 37% over the phone. Alarmingly, 44% say they’ve felt pressured into providing information to AI agents just to complete a task. The discomfort is compounded by perceptions of vulnerability, with 62% believing AI agents are more easily tricked by hackers than humans.
“Agentic AI is changing the way businesses operate, but consumers are clearly signalling a lack of confidence,” said Michael Callahan, CMO at Salt Security. “What many organisations overlook is that the safety and success of AI depends on APIs that power it and they must be effectively discovered, governed and secured. Otherwise, the trust gap will widen, and the risks will escalate.”
APIs form the digital foundation for AI agents, enabling them to retrieve data, trigger actions, and interact across platforms. However, each connection adds a potential attack surface. As AI agents automate more tasks and handle sensitive information, weaknesses in API authentication, input validation and access control become high-risk vulnerabilities.
The Security Shortfall: API Risk Is Underestimated
Despite the growing risk, security practices remain uneven. Only 32% of organisations conduct daily API risk assessments, while 7% report doing so monthly or less. Meanwhile, only 37% say they use dedicated API security solutions, and the same proportion have a dedicated data privacy team overseeing AI initiatives.
The report outlines key best practices, including:
- Continuous API monitoring and anomaly detection using AI tools
- Strong authentication frameworks and least-privilege access
- Encryption for data in transit and at rest
- Regular security testing and developer training
“As AI agents become more autonomous and embedded in business operations, securing the APIs that power them should be an urgent priority as this is a problem that will just keep compounding until it’s out of control,” Michael Callahan concluded. “Securing API infrastructure needs to happen now to reduce risk, improve trust, bolster innovation and increase overall cyber resilience.”
Download the full Salt Security report, Securing the Future of Agentic AI, here to see the complete findings and recommended security actions.
Related News:
Unmask Hidden API Vulnerabilities with Salt Surface from Salt Security
Salt Security: How CISOs Are Approaching API Risk Survey Report
Methodology:Â Censuswide carried out a survey on behalf of Salt Security of 1000 US-based consumers and 250 organizations with 250+ employees who are already using agentic AI
