This week, we explore why AI is shifting inward, from the apps you file taxes with to the platforms that manage enterprise data.

AI in Action

TurboTax’s ChatGPT debut and the shift toward private AI [Enterprise AI/Integration]

If you have ever filed a tax return in TurboTax or tracked expenses in QuickBooks, this week’s biggest AI story may feel surprisingly close. Intuit, the company behind those tools, announced a multi-year partnership worth more than 100 million dollars with OpenAI to bring its financial apps into ChatGPT and expand its use of OpenAI models across TurboTax, Credit Karma, QuickBooks and Mailchimp. ChatGPT users will be able to open Intuit apps inside the chatbot and, with permission, share their financial information for personalized guidance. On the same day, OpenText introduced its new AI Data Platform. OpenText is the company many large organizations use to store, manage and secure their documents and records, and its new platform is designed to let them apply generative tools within their own environment. The system will roll out over an eighteen-month roadmap and support the next wave of the company’s Aviator AI agents.

The details that matter most in both announcements concern privacy and control. Intuit says customer data will remain inside its own environment, protected by the safeguards it already uses for financial information. With user permission, it can apply OpenAI models to financial data inside Intuit’s systems, although neither company has published the full details of how OpenAI processes or retains that information. OpenText frames its platform as a unified governance layer with metadata, access controls, encryption and PII protection. It says customer data is intended to remain in the organization’s chosen environment, though the specifics will depend on each deployment.

Both moves point to the same direction in enterprise AI. Companies with large stores of proprietary data are investing in environments where AI can operate safely on that information rather than sending sensitive data into public tools. Many organizations are likely to follow the same pattern as features roll out. This signals a gradual shift toward workflows in tax, accounting and document management that quietly lean on AI beneath the surface. The question for readers is: how much trust can you place in these governance structures as AI moves deeper into the systems that hold your financial and business information?

Bits of Brilliance

The Gravitational Shift in Enterprise AI [AI Infrastructure/Governance]

A quiet architectural shift seems to be taking place in enterprise AI. For years, the default model was straightforward. Companies sent their data to an AI system, the system returned an answer and the exchange was complete. Inside large organizations, that logic is reversing. Sensitive financial records, contracts and internal documents often carry too much regulatory risk or operational complexity to move freely. Architects sometimes call this “data gravity.” The more important and voluminous the data, the harder it is to relocate. As a result, companies are beginning to bring the model to the data instead. Running AI inside their own governed environments gives them the advantage of existing identity systems, permissions and audit trails while keeping sensitive information under their control.

This shift also improves reliability. When a model works with well-structured inputs, such as tax forms, accounting entries or standardized documents, it is not trying to improvise around missing or inconsistent information. Hallucinations do not disappear, but they become less likely because the data is cleaner, the tasks are defined and the sources are known. At the same time, the vendor landscape is being reshaped. Software and infrastructure providers are no longer judged only by how powerful their models are, but by whether they can offer strong governance, seamless integration and reassuring guarantees that data does not leave the customer’s environment. In this world, governance becomes a competitive advantage rather than a compliance requirement.

For many companies, this means the path to trustworthy AI begins long before the model is selected. Organizations that already understand where their data lives and who owns it will be able to adopt AI more quickly and with fewer surprises. Those with scattered systems or unclear oversight will find that even the best model cannot overcome disorganized information. The lesson is straightforward. Enterprise AI will advance fastest where the data is grounded, the rules are clear and the environment can be trusted. Move the model to the data, strengthen the guardrails around that data and the intelligence that emerges becomes far more reliable and far easier to build on.

_{Byte-Sized Intelligence is a personal newsletter created for educational and informational purposes only. The content reflects the personal views of the author and does not represent the opinions of any employer or affiliated organization. This publication does not offer financial, investment, legal, or professional advice. Any references to tools, technologies, or companies are for illustrative purposes only and do not constitute endorsements. Readers should independently verify any information before acting on it. All AI-generated content or tool usage should be approached critically. Always apply human judgment and discretion when using or interpreting AI outputs.}