Volume 343, Issue 3 · 7 articles
Okta outlines four identity-based practices government agencies can follow to secure agentic AI.
Okta's federal field CTO argues that AI offers government agencies major gains in citizen services, supply chains and data processing, but without a strong security foundation autonomous agents can become unmonitored backdoors to sensitive data.
He frames AI security as an identity problem and points to the Blueprint Alliance, which Okta co-founded to define a reference architecture for securing the agentic enterprise around four fundamental questions.
The post turns those questions into four practices agencies can promote during Cybersecurity Awareness Month.
Okta and CodePath pair identity guardrails for AI agents with training a new generation of cyber defenders.
Okta says AI agent adoption is outpacing governance, citing its AI Agents at Work report finding that 92% of surveyed executives report widespread or moderate use of autonomous agents.
It also cites Gartner's expectation that more than 40% of agentic AI projects will be canceled by the end of 2027 due to cost, unclear value or inadequate risk controls.
The article explains why Okta and CodePath are working together to pair identity guardrails with cybersecurity education that builds the talent needed to secure AI.
Okta adds Client ID Metadata Document support to Okta Identity Engine to simplify registration of AI agents and MCP clients.
Okta says traditional OAuth client registration assumes a small, static set of clients created by admins or via dynamic client registration, which breaks down as AI agents, MCP servers and autonomous workloads are spun up and retired by software at high speed.
The registration step, rather than authentication, has become the bottleneck for agent identity.
Okta has introduced two capabilities, including Client ID Metadata Documents, available to all Okta Identity Engine customers, and explains how they streamline registration without weakening security.
Okta explains how it treats AI agents as first-class identities within its identity security fabric.
Okta says organizations are prioritizing management of non-human identities such as API keys, service accounts, bots and AI agents as generative AI and automation expand the identity attack surface.
Without governance, credential hygiene, monitoring and lifecycle management, these identities pose serious risk.
Okta presents the identity security fabric as a unified framework for consistent posture, governance and visibility across human and machine identities, and describes how it secures AI agents as first-class identities with the same controls and audit trails applied to people.
Okta illustrates how a unified identity platform improves employee productivity and security compared with fragmented tools.
Okta describes modern identity management as a unified approach to securing and streamlining access, authentication and governance across all applications from one platform.
Following a fictional marketing manager through her workday, the post contrasts past friction from repeated passwords, VPNs and slow access requests with single sign-on to everything she needs.
Okta argues many organizations struggle with identity spread across separate SSO, MFA, lifecycle and privileged access tools, and says dedicated identity platforms typically deliver lower total cost of ownership.
Okta's Deputy CSO Charlotte Wylie discusses how AI is accelerating phishing and why phishing-resistant authentication matters.
Okta says phishing tactics have evolved from typo-filled emails to highly personalized, multi-channel attacks, and that AI has accelerated the pace.
Okta Threat Intelligence observed a 136% jump in voice-related phishing attacks in a single month, as attackers use AI to quickly spin up convincing lures.
In this executive conversation, Deputy CSO Charlotte Wylie discusses how organizations can respond with phishing-resistant authentication and security fundamentals.
Okta for AI Agents adds capabilities to discover, register and govern AI agents and their MCP server connections at scale.
Okta says security teams struggle to track AI agents across platforms, record details such as owner and purpose, and govern which MCP servers agents can reach, while AI teams cannot wait on manual registration.
It cites Gartner predictions that large enterprises will run more than 150,000 agents by 2028 while only 13% of organizations believe they have adequate agent governance.
The latest Okta for AI Agents release brings in agents from more builder platforms, adds governance details and cuts manual credential work and per-server setup.