Okta's report shows most enterprises now use multiple AI tools.
Anthropic leads in enterprise adoption, followed by OpenAI.
This shift demands stronger identity and security management across platforms.
What happened
Okta Japan released its Enterprise AI Index on July 17, 2026, analyzing access data from over 2,000 organizations over four years. It found that 109 AI apps and tools are being used across 74 software categories. Among AI-native apps, Anthropic ranked first and OpenAI second, while AI-augmented apps saw Google Workspace third and GitHub fourth. By March 2026, Anthropic and OpenAI overtook Microsoft 365 in enterprise accounts, and by the fourth quarter of 2026, Anthropic led in monthly active users.
Why it matters
The report highlights a shift from single-provider dependence: as of June 2026, about 57% of organizations use two or more AI platforms, while the share using only one provider dropped by 1.2 percentage points in the five months from May to June. This trend suggests enterprises are adopting a 'fit-for-purpose' approach, combining AI tools based on specific needs, rather than relying on a single vendor. It also raises new security challenges as AI agents become more autonomous and require identity management across multiple platforms.
What to watch
Okta Japan emphasizes that as AI agents become more prevalent, organizations will need to address three security questions: 'where are they?', 'what can they connect to?', and 'what can they do?' — underlining the growing importance of identity-centric security in a multi-provider AI environment.
Ask the AI about this article →
Okta Japan's Enterprise AI Index provides a data-driven snapshot of how AI tools are being adopted inside companies, based on four years of access data from over 2,000 organizations. The key shift is the rise of "AI-native" apps—those built specifically for AI, like Anthropic's Claude and OpenAI's ChatGPT—over "AI-augmented" suites like Microsoft 365 and Google Workspace. By March 2026, Anthropic and OpenAI had overtaken Microsoft 365 in enterprise accounts, and by the fourth quarter of 2026, Anthropic led in monthly active users. This marks a clear change from earlier periods when ChatGPT was the default choice.
This trend toward multi-platform use—57% of organizations now use two or more AI platforms—reflects a pragmatic, fit-for-purpose approach rather than a single-vendor strategy. However, it also introduces new complexity: IT and security teams must manage identities across multiple AI agents, each with its own access permissions and potential security risks. Okta Japan's emphasis on identity-centric security suggests that as AI agents become more autonomous, the ability to govern who and what they can access will be critical. For business readers, this means that AI adoption is no longer just about picking the best tool—it's about ensuring that the entire ecosystem can be managed securely.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
On August 25, 2026, OpenAI unveiled Jalapeño, its first custom inference chip, claiming 1.5–1.9x more AI work…
The Wall Street Journal published an op-ed by billionaire investor Stanley Druckenmiller that was written with…
Lovable introduced a vision where apps become 'capabilities' that AI agents can call directly, bypassing the h…
OpenAI CEO Sam Altman told TIME the company will have an internal system he'd call AGI by the end of 2026
Scalable Capital, a Munich-based digital investment platform with over €60 billion in client assets, has becom…
Cigna projects its AI and predictive analytics tools, used to identify chronic conditions like cancer and kidn…
