Technologies
OpenAI
Frontier models, wired into real work.
OpenAI moves fast, and last year’s integration advice is already out of date. We design, build and run OpenAI solutions with the governance your organisation needs, from ChatGPT rollouts to production API workloads.
YOUR OPENAI ESTATE · MAPPED
EVERYDAY AI
ChatGPT Business
ChatGPT Enterprise
THE PLATFORM
GPT-5 API
Realtime API
Embeddings API
Responses API
AGENTS AND CODING
Agents SDK
Codex
IMAGES AND SPEECH
GPT Image
Whisper
10 products · 4 layers
governed and adopted, not just licensed
Products and services
What we deliver across the OpenAI stack.
Twelve products, four layers. Start with the layer where your pain lives.
Everyday AI
Where your people meet OpenAI
ChatGPT Business
The team workspace for ChatGPT: shared workspaces, admin controls and company data excluded from training by default.
ChatGPT Enterprise
ChatGPT for large organisations: unlimited access to frontier models, SSO, compliance tooling and enterprise support.
The platform
Where your products get their intelligence
GPT-5 API
OpenAI’s flagship model family via API, from fast everyday models to deep reasoning.
Realtime API
Low latency voice and audio conversations for phone agents and live assistants.
Embeddings API
Turns your documents into vectors for search, retrieval and recommendations.
Responses API
OpenAI’s unified API for agents and assistants, the successor to the retired Assistants API. We build on it and migrate to it.
Agents and coding
Where the work gets delegated
Agents SDK
The open source framework for building and orchestrating multi agent workflows on OpenAI models.
Codex
OpenAI’s coding agent: it writes, tests and reviews code across whole repositories.
Frontier
Run agents in production against your systems of record, with the approvals and audit trail a business needs.
AgentKit
Design and version agent workflows on a canvas, and administer every connected data source from one place.
Images and speech
Where words become media
GPT Image
OpenAI’s image generation models, the replacement for DALL-E, available in ChatGPT and through the API.
Whisper
Open source speech to text, with newer OpenAI transcription models available through the API.
Why CG TECH
Why CG TECH for your OpenAI work.
We do not stop at API keys. We design, build and run OpenAI solutions, then train your people so the investment lands.
Governance first
Every rollout comes with identity, policy and audit set up from day one, so your security team can say yes.
We track the churn
OpenAI retires models and APIs on a schedule. We watch every deprecation so your integrations never break by surprise.
Build and run, not just advise
We design the solution, build it in your environment and run it with you until your team owns it.
Adoption that sticks
Training and change support are part of every engagement, because a platform nobody uses is a cost, not an asset.
Where OpenAI fits.
Frontier capability and the fastest moving ecosystem in AI. Strong for voice agents, coding agents and products that need the newest models first.
Where it does not.
If your organisation runs on Microsoft 365 and wants AI inside the tools and governance you already have, Copilot usually lands faster. For long documents and careful writing, Claude is often the better fit. You get one honest recommendation, not a default.
What good looks like
Real projects, real outcomes.
Questions we hear a lot
Common questions about OpenAI
What does CG TECH do with OpenAI?
We design, build and run OpenAI work for Australian organisations, from rolling out ChatGPT Business or Enterprise across a team, through to production workloads on the API: agents, voice, search, transcription and image generation. The governance and cost control come as part of it rather than later.
We already pay for Copilot. Why would we use OpenAI as well?
They do different jobs. Copilot lives inside Microsoft 365. OpenAI's platform powers custom products, voice agents and coding agents. Many organisations run both, each where it is strongest.
Is our data used to train OpenAI models?
On ChatGPT Business and Enterprise plans, company data is excluded from training by default, and API data is not used for training. Consumer plans work differently. We set the right plan and settings during discovery.
Which ChatGPT plan do we need?
Business suits most teams. Enterprise adds unlimited access, deeper compliance tooling and enterprise support. We confirm the right fit during discovery.
We built on the Assistants API. What now?
OpenAI shuts it down on 26 August 2026. New work belongs on the Responses API, and existing integrations need to migrate. We scope and run those migrations.
Do you work with mid-sized businesses?
Yes. Our sweet spot is Australian organisations of roughly <strong>50 to 500 seats</strong>, including not-for-profits and government suppliers of that size. That is small enough that a rollout can be done properly and large enough that getting it wrong is expensive.
Can we use OpenAI and Microsoft together?
Almost every client does. Copilot sits inside Microsoft 365 where your documents live, and OpenAI covers the work that falls outside it or needs a different model. We would rather help you use both well than argue for one.
Where do most OpenAI rollouts go wrong?
Buying seats without deciding what they are for. Licences get handed out, a few enthusiasts use them properly, everyone else opens it twice and forgets. We start with the specific jobs it should do, then size the rollout to that.
Ready when you are
Working with OpenAI? Let us talk.
A discovery session maps your use cases, your risks and your quick wins. You keep the plan either way.
What to expect
- A consultant replies within 4 business hours
- Session booked to understand your requirements
- We will provide you with a fixed price quote