Free, paid or enterprise: Choosing artificial intelligence (AI) with confidence

AI tools look similar on the surface, but their interpretation of prompts, files and data can be very different.

A useful way to understand the market is to group AI services into three broad categories:

• Public (free)
• Private (paid consumer)
• Enterprise (corporate)

How these services are implemented determines who can access them, how submitted information may be used and what contractual, security and administrative controls are provided, if any.

Public AI (the ‘free’ tier, generally available)

Public AI is a broadly available service that anyone can access, often without an organisational agreement. It is convenient for general research, brainstorming and drafting, but it offers minimal security control.

Prompts and uploaded content (or ‘input’) leave the user’s managed environment and are processed on the provider’s platform. Depending on the service, account type and settings, that input may be retained, reviewed or used to improve future models.

Public AI should therefore be treated like any external online service: confidential client information, personal data, passwords, financial records and commercially sensitive material should not be entered unless the organisation has explicitly approved the tool and its terms.

Private AI (the ‘paid’ tier, for consumers)

Private AI, also known as ‘account-based AI’, requires users to authenticate with the service and receive a personal workspace, conversation history or enhanced privacy settings. However, ‘private’ does not necessarily mean that the input data is excluded from model training.

Some providers may still use prompts and responses to improve their services unless the user manually changes a setting, opts out, or purchases a plan with stronger contractual protections. This category can feel private because conversations are not publicly visible, yet the provider may still retain and process the content under its consumer terms.

Users should read the applicable privacy policy, retention rules and training controls rather than relying on the product label alone.

Enterprise AI (the ‘corporate’ tier, for business and organisations)

Enterprise AI is designed for organisational use under a business agreement. In a properly configured enterprise service, input, output and organisational data are not used to train the provider’s general-purpose models.

Data is protected through enterprise identity, access controls, encryption, audit capabilities, retention settings and administrative policies. These services may also support compliance obligations, data residency requirements and integration with approved business systems.

Enterprise AI is therefore generally better suited to work involving confidential, regulated or commercially sensitive information. It is not risk-free, however: security still depends on correct licensing, configuration, permissions, information classification and user behaviour.

Choosing the right AI for your business

The safest approach is to match the tool to the sensitivity of the task.

For organisations already using Microsoft 365, Microsoft Copilot can provide an enterprise-focused option. It works within familiar applications such as Word, Outlook, Excel, PowerPoint and Teams and, where applicable, respects existing Microsoft 365 identities and permissions.

This can help organisations use AI within their existing technology environment while retaining organisational controls around areas such as access, information protection, retention and administration. It can also help staff accelerate tasks such as drafting, summarising, analysing information, preparing presentations and communicating without needing to copy business information into an unmanaged public service.

Regardless of the AI platform selected, organisations should configure permissions carefully and train users to verify AI-generated output. Products should also be assessed based on factors such as data use, retention, data location, compliance and integration, rather than simply the quality of the AI model.

AI can create significant efficiencies, but those benefits need to be balanced with appropriate safeguards around your organisation’s information. As AI becomes part of everyday business, now is a good time to review which tools your people are using, what information is being shared with them and whether the right governance and protections are in place.

Choosing AI with confidence

Not sure whether your current AI tools and practices are appropriate for your business?

Contact Andrew Grgic or your trusted Nexia Advisor about how you are using AI today and the steps you can take to use it more securely and confidently.